Breakout-Strategie Breakout-Trading Technique Artikel-Sammlungen: BASIC und ADVANCED. Ausbruch-Strategie für Forex Trading Technische Tool-Erkenntnis: Preisausbruch BY ACTIVE TRADER STAFF (Active Trader, März 2001) 2 5 9 Mehr bang for your buck: Muster in Mustern DURCH ACTIVE TRADER STAFF (aktiver Trader, Oktober 2000) Brechen und Rutschen antizipieren VON STEVE WENDLANDT (Active Trader, August 2000). Trading System Lab: 100-20 Kanal Ausbruch System DURCH DION KURCZEK (Active Trader, Juni 2003). 13 16 18 20 23 28 32 35 39 41 42 47 53 58 60 62 64 66 1 Futures Trading System Lab: 100-20 Kanal Ausbruch System DURCH DION KURCZEK (Active Trader, Juni 2003). Futures Trading System Lab: 60-minütiges Breakout-System von VOLKER KNAPP (Active Trader, Januar 2004). Futures Trading System Lab: Vier-Prozent-Breakout-System von VOLKER KNAPP (Active Trader, September 2004). Ausbreitungsmuster: Hinweise zur Ausbruchsrichtung von THOMAS N. BULKOWSKI (Active Trader, April 2004). BY KEN CALHOUN (aktiver Trader, September 2001) Swing Trading 10-Tage-Channel-Ausbrüche BY KEN CALHOUN (aktiver Trader, im März 2004) Zwei-Minuten-Ausbrüche beherrschen BY THOMAS N. BULKOWSKI 2002). Trading System Lab: Volatilitätsausbruch durch THOMAS STRIDSMAN (Active Trader, Oktober 2002). Futures-Trading-System-Lab: Futures-Volatilitäts-Ausbruchssystem von THOMAS STRIDSMAN (Active Trader, Oktober 2002). Besserer Breakout-Handel: Das Rauschkanalsystem von DENNIS MEYERS, PH. D. (Aktiver Händler, September 2001). Die lange und kurze Zeit: Das Noise Channel Breakout System 2 DURCH DENNIS MEYERS, PH. D (Active Trader, Oktober 2001). Trading System Lab: DeMark Variation BY THOMAS STRIDSMAN (Active Trader, September 2001) Trading System Lab: Dynamisches Ausbruchsystem von THOMAS STRIDSMAN (Aktiver Trader, Februar 2004) 2003). Futures Trading System Lab: Dynamisches Ausbruchsystem von THOMAS STRIDSMAN (Active Trader, Februar 2003). Futures Trading System Lab: Experimentieren mit Exits Von VOLKER KNAPP (Active Trader, Juni 2004). Futures Trading System Lab: Monatlicher Ausbruch BY DION KURCZEK UND VOLKER KNAPP (Active Trader, März 2004). Trading System Lab: 60-minütiges Breakout-System von VOLKER KNAPP (Active Trader, Januar 2004). ACTIVE TRADER activetradermag 68 TRADING Grundlagen Einblick in das technische Werkzeug: Der Preisausbruch gewann seine Wirksamkeit so weit, dass viele Händler auf der Suche nach falschen Breakouts (wenn der Preis durch eine Breakout - Richtung des Anfangsausbruchs (als Ausbleichen des Ausbrechens bezeichnet). Breakouts sind nicht ausschließlich darauf beschränkt, auf neue Höhen einer bestimmten Anzahl von Stäben (d. h. 10-bar, 20-bar oder 40-bar-Ausbrüche) zu bewegen. Wie erwähnt, kann der Preis auch durch die Unterstützung und Widerstand Ebenen der Handelsbereiche oder andere technische Meilensteine wie langjährige Höhen oder Tiefen brechen. Abbildung 1 zeigt 40-Tage-Breakout-Levels auf einem Tages-Chart. Abbildung 2 zeigt die 20-stufigen Breakout-Werte auf einer 10-minütigen Tabelle. Fig. 3 zeigt einen Bruch über dem Widerstandswert, der durch eine vergangene signifikante Höhe definiert ist. Preisausbrüche sind die Basis für viele der erfolgreichsten Handelsansätze. Wir erklären die Grundlagen dieser Handelstechnik. Der Preisausbruch ist eines der einfachsten und mächtigsten Konzepte im Handel. Es tritt auf, wenn sich der Preis aus einer Konsolidierungs - oder Handelsspanne herauszieht (eine Periode relativ schmaler, seitwärts gerichteter Kursbewegungen) oder über oder unter ein etabliertes Preisniveau (Unterstützung oder Widerstand) schreitet, was entweder einen vorübergehenden oder einen anhaltenden Trend zur Folge hat. Die Tendenz, auf neue Höhen oder Tiefen zu drängen (vor allem, wenn das Preisniveau in der Vergangenheit wiederholt getestet wurde) ist ein Beweis für eine starke Dynamik und deutet darauf hin, dass der Markt das Potenzial hat, in dieser Richtung fortzufahren. Mit anderen Worten, die Grundlogik hinter den Preisausbrüchen besteht darin, dass ein Markt, der neue Höchststände macht (und mit einem weiteren Preiszuwachs), Stärke ausstellt und gekauft werden sollte, während ein Markt neue Tiefststände macht (und mit einem weiteren Preisrückgang) Schwäche aufweisen und verkauft werden sollten. Zum Beispiel ist der Grund, warum neue 52-Wochen-Hochs oder Tiefs in Aktien so häufig referenziert ist, weil die implizite Bedeutung der Preis brechen durch diese Ebenen. Dieses Konzept der Preisbewegung gilt sowohl für Intraday-Zeitrahmen als auch für tägliche oder monatliche. Die vierwöchigen Hochs oder Tiefs repräsentieren einfach nur natürliche Widerstands - und Unterstützungsniveaus. Diese Art von Trading-System wird oft als Stop-and-Reverse (SAR) bezeichnet, weil, wenn ein Handelssignal erzeugt wird, die bestehende Position liquidiert (gestoppt) und eine neue Position (eine umgekehrte der vorherigen) hergestellt wird . Diese grundlegende Handelsregel, die weit verbreitete Popularität als der 20-Tage-Ausbruch gewann, war ein integraler Bestandteil vieler populärer mechanisierter Handelsstrategien, am berühmtesten die von Futures Trader Richard Dennis Gruppe von Trend-Anhänger wie die Schildkröten bekannt. Trendfolgende Trader (vor allem in den Futures-Märkten) nutzten diese einfache Technik oder eine Variante davon, um starke Trends in den 1970er und 80er Jahren zu nutzen. Jedoch hat die weit verbreitete Beliebtheit des 20-tägigen Breakout-Levels weniger als 40 Tage dämmrige Breakout-Werte, sowohl hoch als auch niedrig. Eine grundlegende Ausbruch-Ansatz ist zu kaufen, wenn der Preis übersteigt die n-bar (in diesem Fall die 40-Tage) hoch und verkaufen, wenn es unter dem n-bar niedrig. Höchster Preis der letzten 40 Bar Oracle Corporation (ORCL), täglich 45 40 37.00 35 Donchian Ausbruch Ebenen Der Begriff Ausbruch wird oft mit Richard Donchian, die erste Person, die systematische Nutzung der Ausbruch Ebenen zu popularisieren. Seine Grundannäherung wurde die Donchian-Vierwochenregel genannt, die aus folgendem bestand: 1. Gehen Sie lange (und Deckungs-Short-Positionen), wenn der Markt eine neue Vierwoche hoch macht (dh wenn der Preis den höchsten Preis der vorherigen vier übersteigt Wochen). 2. Gehen Sie kurz (und decken Sie lange Positionen), wenn der Markt eine neue vier Wochen niedrig macht (dh, wenn der Preis unter den niedrigsten Preis der letzten vier Wochen sinkt). 2 30 26 916 25 Niedrigster Preis der letzten 40 Tafeln 21,50 20 15 27 3 10 24 31 7 14 28 6 13 20 27 3 10 24 1 8 15 22 30 5 12 19 26 3 10 17 24 31 7 14 21 28 5 11 18 25 2 9 16 23 30 6 13 27 4 Jan. 2000 Feb. Mär. Apr. Mai Juni Juli Aug. Sept. Okt. Nov. Dez. Quelle: QCharts nach Zitat activetradermag März 2001 ACTIVE TRADER Glossar ABBILDUNG 2 DONCHIAN BREAKOUT CHANNELS, INTRADAY Das Breakout-Konzept ist auf jeden Zeitrahmen anwendbar. Hier werden die höchsten 20-bar-Höhen und die niedrigsten 20-Balken-Tiefen durch die Kanallinien angezeigt. Oracle Corporation (ORCL), 10 Minuten 27,25 27 26 916 26 Höchster Preis der letzten 20 Takte 25 24 23,81 23 Ein falscher Breakout tritt auf, wenn der Preis in der erwarteten Richtung durch einen Unterstützungs - oder Widerstandswert dringt, was auf einen neuen Preisschub oder Trend hinweist, Nur bis (relativ) schnell umgekehrte Richtung, wenn kein reales followthrough materialisiert. Denn Händler, die gekauft oder verkauft auf den ersten Ausbruch können alle klettern auf einmal, um aus ihrem Gewerbe, wenn der Markt nicht durch zu folgen, die Umkehr kann sehr kraftvoll sein. Aus diesem Grund verblüffen konträre Händler manchmal erste Ausbrüche, um diese kurzfristigen Umkehrungen zu aktivieren. Stop-and-Reverse (SAR) bezieht sich auf einen Trading-Ansatz, der immer auf dem Markt, lang oder kurz ist. Die bestehende Position wird aufgelöst (gestoppt) und eine neue Position (eine Umkehrung der vorherigen) wird hergestellt, wobei das gleiche Signal in die entgegengesetzte Richtung verwendet wird. Zum Beispiel würde ein einfaches 40-Tage-SAR-Breakout-System kaufen, wenn der Preis den höchsten Höchststand der letzten 40 Tage übersteigt und verkauft, wenn der Preis unter das niedrigste Tief der letzten 40 Tage fällt. Unterstützung und Widerstand. Support ist ein Preisniveau, das als Boden fungiert und verhindert, dass die Preise unter diesen Wert fallen. Widerstand ist das Gegenteil: ein Preisniveau, das als Obergrenze fungiert, eine Barriere, die verhindert, dass die Preise steigen. Niedrigster Preis der letzten 20 Bars 14 15 10 11 12 11/28 Dienstag 13 14 15 10 11 12 11/29 Mittwoch 13 14 15 10 11 12 11/30 Donnerstag 13 14 15 22 10 12/1 Freitag Quelle: QCharts nach Zitat Die Donchian-Breakout wird auch allgemein als Preiskanal Ausbruch bezeichnet. Anwendung Händler, die Ausbrüche verwenden, beruhen auf folgenden Grundsätzen: Wenn die Preisdynamik stark genug ist (entweder nach oben oder unten), um eine bedeutende technische Ebene durchzusetzen, besteht eine gute Chance, dass sich der Preis in dieser Richtung zumindest für eine Weile fortsetzt. Infolgedessen stellen diese Preisniveaus logische Handelsein - und - ausstiegsniveaus mit gut definiertem Risiko dar, sowohl für Händler, die erwarten, in Richtung des Ausbruchs zu folgen, und, wie kurz beschrieben wird, Händler, die Ausbrüche brechen wollen. Können Händler, die lange auf dem Breakout gehen, Schutzstopps in einer Reihe von technisch logischen Orten, in Bezug auf die Reichweite. Erstens könnte der Stopp unterhalb des Niedrigwerts der Handelsspanne liegen. Zweitens wäre eine konservativere Stop-Platzierung in der Mitte der Handelsspanne (oder in den oberen 25 Prozent der Handelsspanne usw.). Schließlich ist die konservativste Alternative ein Stopp direkt unterhalb des ursprünglichen Ausbruchpegels, der in Fig. 3 verwendet werden könnte. BREAKOUT VORHER HOCH Ein vorheriges Hoch erzeugt ein Widerstandsniveau, das mehrmals getestet wird, bevor der Preis nach oben ausfällt. Eine signifikante Trendbewegung folgt. Sun Microsystems Inc. (SUNW), wöchentlich 28 25 5964 24 Wichtige Punkte Die Preisausbrüche werden typischerweise als Trendfolgesignale verwendet. Je größer die Anzahl der Tage (oder Preisstäbe), die verwendet werden, um den Ausbruch zu bestimmen, wird der längerfristige Trend das Handelssystem reflektieren und versuchen zu nutzen. Zum Beispiel würde ein 20-tägiger (oder 20-bar) Ausbruch kürzere Trends erfassen als ein 40-tägiger Ausbruch, der wiederum kürzere Trends widerspiegeln würde als ein 80-Tage-Ausbruch. Im Allgemeinen, in Bezug auf Trend-Ansatz, je länger der Ausbruch, desto größer die Preisbewegung und desto größer die Wahrscheinlichkeit einer nachhaltigen folgen. Breakout-Handel kann auch die Risikosteuerung vereinfachen, da Stop-Loss-Level oft leicht zu identifizieren sind. Zum Beispiel, wenn der Preis bricht aus der Oberseite eines trad - Breakout oben getesteten hohen 20 16 12 8 Jan. 1997 Apr. Juli Okt. Jan. 1998 Apr. Juli Okt. Jan. 1999 Quelle: QCharts nach Zitat ACTIVE TRADER März 2001 activetradermag 3 ABBILDUNG 4 TRADING RANGE BREAKOUT MIT STOP LEVELS Die Grenzen einer Handelsspanne bieten logische Stop-Level für einen Breakout-Handel. Nach einem Abbruch der Bandbreite könnte ein Trader, je nachdem wie konservativ er war, eine Stop-Loss-Order auf der ursprünglichen Breakout-Ebene, dem Mittelpunkt des Bereichs (oder einem anderen Punkt innerhalb des Bereichs) oder der oberen Ebene von platzieren die Reichweite. American Express Inc. (AXP), 2 Minuten Weite Seite des Handelsbereichs (Haltestelle 1) Mittelpunkt (Haltestelle 2) 54 55 53 Handelsbereich Breakout-Ebene (Haltestelle 3) 51 1516 51 10:00 10:30 11:00 11: 30 12:00 12:30 13:00 13:30 14:00 14:30 15:00 15:30 9:30 10:00 10:30 11:00 11/21 Dienstag Handelsspanne und mögliche Stopppunkte. Abbildung 5 zeigt die umgekehrte Situation. Die Aktie bricht aus, um den Nachteil der Handelsspanne, aber dies stellt sich heraus, ein falscher Ausbruch sein. Die Aktie kehrt zurück in die Handelsspanne und bricht schließlich durch die Oberseite der Handelsspanne aus. Wiederum stellen die Grenzen (und der Mittelpunkt) des Handelsbereichs logische Stop-Pegel sowohl für den anfänglichen Abwärts-Breakout als auch für den anschließenden Upside-Break bereit. Wegen der Möglichkeit von falschen Bewegungen bei beliebten Ausbruch Ebenen, Händler suchen, um Trends zu erfassen manchmal verwenden Bestätigungssignale, um die Wahrscheinlichkeit des Erfolgs zu verbessern. Zum Beispiel kann der Händler nach einem anfänglichen Aufwärtsausbruch darauf warten, dass der Markt für eine bestimmte Anzahl von Balken über dem Break-Level (oder darüber darüber) verbleibt oder ihn um einen bestimmten Prozentsatz durchdringt. Solche Techniken verzögern den Eintritt und begrenzen das Gewinnpotential (und führen zu einigen verpassten Trades), können aber auch falsche Signale reduzieren. Quelle: QCharts nach Quote Bottom line Das Breakout-Konzept ist eines der wichtigsten im technischen Handel. Der Kauf von Märkten mit starken Aufwärtsbewegungen mit weiteren Potenzialen für die Aufwärtsbewegung und der Absatz von Schwellenmärkten (Downside Breakouts) mit weiteren Potenzialen für Abwärtsbewegungen sind die Basis vieler Handelspläne und - systeme auf vielen Zeitrahmen. In ähnlicher Weise sind falsche Ausbrüche die Grundlage für einige gegenläufige Handelstechniken. Das Breakout-Konzept ist auch für Händler interessant, die an einem systematischen Ansatz interessiert sind. Ein sehr kurzfristiger Händler. Alle diese Entscheidungen haben eines gemeinsam: Die Platzierung des Anschlags entspricht einer Kursbewegung, die die Gültigkeit (in unterschiedlichem Ausmaß) des ursprünglichen Ausbruchs negiert. Wenn der ursprüngliche Grund für einen Handel aufgehoben wird, sollte diese Position beseitigt werden. (Anmerkung auch, die zweite und dritte Option wären wahrscheinlich kurze Einstiegspunkte für Händler, die den Upside-Break ausblenden möchten). Abbildung 4 zeigt einen Downside-Breakout aus einer Figur 5 FALSE BREAKOUT UND REVERSAL In diesem Fall bricht die Aktie zunächst aus Der Boden der Handelsspanne, nur um zurückzukehren zurück in den Handelsbereich und schließlich brechen aus dem oberen Bereich des Bereichs. In beiden Fällen werden die Stop-Loss-Werte wieder leicht identifiziert. Microsoft Corporation (MSFT), täglich 38 36 35 2364 34 32 Zusätzliche Forschung: Handel für ein Leben von Alexander Elder John Wiley Sons, 1993 Handelssysteme und Methoden von Perry Kaufman 3. Auflage, John Wiley Sons, 1998 Technische Analyse der Finanzmärkte von John Murphy New Yorker Institut für Finanzen, 1999 Street Smarts von Linda Raschke und Laurence A. Connors M. Gordon Verlagsgruppe, 1995 Schwager über Futures: Technische Analyse von Jack Schwager John Wiley Sons, 1996 Weite Seite der Handelsspanne Handelsbereich 30 28 26 Mittelpunkt 24 22 20 Falscher Durchbruch Ursprünglicher Durchbruch 25 2 9 16 23 30 6 13 20 27 3 10 18 24 3 10 17 24 31 7 14 21 28 5 12 19 27 2 9 16 23 30 7 14 21 Dez. Jan. 1997 Feb März 04 ACTIVE TRADER TRADING Strategien Mehr bang for your buck: MUSTER INNERHALB MUSTER Hut macht einen guten Trade Nun, im Rückblick würden die meisten Händler sagen, ein schöner Gewinn macht einen guten Handel. Aber wenn youre setzen eine Position auf, das Ergebnis ist unvorhersehbar. Wed alle gerne wissen, ein Handel wird gut im Voraus, aber leider sind die Märkte nicht so entgegenkommend. Was Sie suchen, wenn youre, das in einen Handel erhält, ein Einstiegpunkt ist, in dem die Vorteile einer Bewegung zu Ihren Gunsten besser als Durchschnitt sind. Dann, indem Sie einen Plan 1 FALSE BREAKOUT Eine Handelsspanne entwickelt sich im Anschluss an eine scharfe Rallye. Nach einem anfänglichen Aufwärtsausbruch, kehrt die Aktie auf den Nachteil, stoppt die lange Position. Microsoft Corporation (MSFT), täglich Upside-Ausbruch 82 80 78 76 W Stützpegel als Anfangsstopp verwendet Stopped out 74 72 58 72 70 68 12 19 26 3 Juli 10 17 24 31 Aug. 7 Quelle: Qcharts nach Zitat, die bestimmt, wann und wo Youll Ausgang mit entweder einem Verlust oder einem Gewinn, versuchen Sie, einen Handel, wo die potenzielle Belohnung ist größer als das bekannte Risiko zu strukturieren. Der Vorteil von Handelsausbrüchen von Stauungsmustern wie Handelsstrecken, Dreiecken, Flaggen und Wimpeln ist, dass diese Formationen Ihnen erlauben, das Risiko auf Ihren Berufen klar zu definieren. Zum Beispiel, wenn eine Aktie bewegt sich in eine Handelsspanne nach einer Rallye, können Sie schauen, um ein Upside Breakout des Bereichs in Erwartung einer Fortsetzung des Aufwärtstrends zu kaufen. Der logische Platz, um einen anfänglichen Schutzstopp zu setzen, liegt unterhalb des Niedrigwerts der Handelsspanne, da eine Nachteilumkehr durch die Unterstützung der Reichweite eine bärische Entwicklung wäre. Abbildung 1 zeigt ein Beispiel. Ende Juni stellte Microsoft (MSFT) nach etwa einer 16-Punkte-Rallye eine relativ enge Handelsspanne auf. Am 6. Juli brach die Aktie am Rande der Reichweite (rund 80 18) aus. Der erste Schutzstopp wäre knapp unter dem Stützniveau der Handelsspanne, etwa 76 12, gelegen Aufwärtstrend war eigentlich ein falscher Ausbruch und der Handel sollte aufgegeben werden. Das ist genau das, was passiert ist. Zwei Tage nach der Einfahrt hatte sich die Aktie wieder in die Handelsspanne zurückgezogen. Es bewegte sich seitwärts, um über die folgenden einige Tage vorher zu senken, am 19. Juli, der den Nachteil der Strecke eindringt und den langen Handel stoppt. Das Risiko für diesen Handel betrug moderat 3 58 Punkte. Aber was tun Sie, wenn eine Handelsspanne viel breiter ist und ein Stopp auf der Basis der Unterstützung oder des Widerstandes ein zu großes Risiko darstellt? Abbildung 2 5 activetradermag Oktober 2000 AKTIVER TRADER Wie man Handelsgelegenheiten mit erhöhter Belohnung und verringertem Risiko durch Handelsmuster schafft Innerhalb von Mustern. ABBILDUNG 2 BEREICHSRISIKEN Die Verwendung der gegenüberliegenden Seite eines Handelsbereichs als Stopp für einen Breakout-Handel kann zu einem großen anfänglichen Risiko führen, wenn die Handelsreichweite breit ist. Die International Business Machine Corp. (IBM), täglich 130 125 120 1516 120 115, zeigt eine weitaus volatilere Handelsspanne als diejenige in Abbildung 1. Nach dem gleichen Ansatz wie im vorigen Beispiel, Würde ein anfänglicher Schutzstopp unterhalb des Niedrigwerts des Bereichs ein beträchtliches Risiko darstellen. Infolgedessen setzen einige Händler den Anfangsstopp in der Mitte der Handelsstrecke ein. Diese konservativere Methode beruht auf der Vorstellung, dass ein starker Breakout-Kurs sofort eintreten und nicht wieder in den Handel eindringen sollte. 110 105 100 95 90 4 11 18 25 1 8 15 29 6 13 27 3 10 24 31 7 14 28 6 13 20 27 3 10 24 1 8 15 22 30 5 12 19 26 3 10 17 24 31 7 14 Nov. Dez. Jan. 2000 Feb. Mar. Apr. Mai Juni Juli Aug. Quelle: Qcharts nach Zitat ABBILDUNG 3 CONGESTION WITHIN CONGESTION A Kürzere, schmalere Handelsbereich bildet sich nur auf dem Widerstandswert eines größeren Bereichs. Die Verwendung des Stützpegels des kleineren Bereichs als Schutzniveau für einen Aufwärtsausbruch verringert das Anfahrrisiko erheblich. Oracle Corporation (ORCL), täglich 90 Reichweite. Ein weiterer Weg, um das Risiko auf Breakout Trades zu reduzieren, ist es, kürzere Muster in größeren Mustern zu suchen, mit denen Sie Ihren anfänglichen Stop-Loss näher an Ihren Einstiegspunkt legen können. Muster innerhalb von Mustern Wenn das Risiko, das von einer bestimmten Handelsspanne impliziert wird, außergewöhnlich groß ist, können Sie nach kleineren Überlastungsmustern in der Nähe der Unterstützungs - oder Widerstandsebenen des Bereichs suchen. Die Einstiegs - und Stopppunkte auf den durch das kleinere Muster definierten Ebenen können das Risiko für den Handel reduzieren und die Möglichkeit bieten, frühzeitig in die Position zu gelangen. Abbildung 3 zeigt die Bildung einer breiten Handelsspanne in Oracle (ORCL) zu Beginn dieses Jahres. Ein Trader, der lange auf einen Aufwärtsausbruch dieser Strecke eintreten möchte, müsste ein Risiko von mehr als 16 Punkten annehmen, vorausgesetzt, dass der untere Bereich für den anfänglichen Stop-Loss verwendet wurde. Allerdings entwickelte sich im Februar eine viel schmalere Handelsspanne. Mit dieser Palette als Basis für einen Up-Breakout-Handel hätte der gleiche Eintrag 6 80 77 Größere Handelsspanne 70/1000 5 Verträge angeboten. Der 5.000 maximale Verlust dieses Fünf-Vertrags-Handels sollte nicht mehr als 2 Prozent des aktuellen Portfolio-Equity. Infolgedessen wäre das System nicht in der Lage, diese Position zu übernehmen, wenn das Kontoguthaben nicht mehr als 250.000 beträgt. Eigenkapital: 100.000. Deduct 10 Schlupf / Provision pro Handel. Testdaten: Das System wurde auf dem Active Trader Standard-Futures-Portfolio getestet, das die folgenden 20 Futures enthält: DAX30 (AX), Mais (C), Rohöl (CL), Deutscher Bund (DT), Eurodollar (ED), Euro (LH), Nasdaq 100 (ND), Erdgas (NG), Gold (GC), Kupfer (HG), Japanischer Yen (JY), Kaffee (KC), Lebendschweine (S), Silber (SI), SP 500 16 Linear reg Kurz ACTIVE TRADER Juni 2003 activetradermag ABBILDUNG 3 DRAWDOWN CURVE: 2 PERCENT MAXIMALES RISIKO Der maximale Drawdown betrug etwa 18 Prozent. 0,00 -2,00 -4,00 -6,00 -8,00 -10,00 -12,00 -14,00 -16,00 -18,00 3/25/93 3/1/94 2/1/95 1/9/96 1/2/97 1/2/98 1 / 4/99 1/3/00 1/2/01 1/2/02 cent. Die Systemergebnisse des Futures-Portfolios waren recht gut, wenn das maximale Risiko auf 2 Prozent pro Trade festgelegt wurde. Das System erzielte einen durchschnittlichen Gewinn von 8,42 Prozent pro Jahr, wobei das größte Verlängerungsjahr -8,91 Prozent betrug. Das Marktrisiko des Systems lag im Durchschnitt bei etwa 30 Prozent. Auf der Grundlage dieser Informationen könnte die Idee, das Risiko zu erhöhen und mehr Verträge für jedes Signal zu nehmen, wie eine gute Idee klingen, zumal es immer noch genügend Spielraum gibt. Obwohl das System einen Kontobetrag von mehr als 3 Millionen erreicht hat (siehe SP) und 10 Jahre T-Notes (TY) Der verwendete Test Verwendet Adaptierte Daten von Pinnacle Data Corp ABBILDUNG 4 DRAWDOWN CURVE: 6 PERCENT MAXIMAL RISIKO Der Drawdown erhöhte sich In der Tiefe und Länge in dieser Version des Systems 0.00 Testzeitraum: August 1993 bis November 2002. -5.00 System-Ergebnisse: Sowohl die langen und kurzen Seiten der sys-10.00 tem waren rentabel, und das Verhältnis von gewinnen zu verlieren -15,00 (Abb. 1) mit einer 2-prozentigen maximalen Verlusteinstellung zeigt einen rela-20.00-positiven reibungslosen, stetigen Aufwärtstrend. Die 6-prozentige maximale -25.00 Verlust-Version (Abbildung 2) Beiträge ein viel größerer Gewinn, Mit -30,00 deutlich höherer Volatilität -35,00 Die Positionierungsmethode hält das System von -40,00 vielen riskanten Positionen aus, obwohl es in einigen Märkten keinen Handel mit sig-45,00 Nals ergab, weil das Risiko während des gesamten Testzeitraums zu hoch war. -50.00 Um die Wirkung der Höhe des Risikos zu veranschaulichen, vergleichen Sie 3/25/93 die Einzugskurven in den Abbildungen 3 und 4. Abbildung 3 ist die Einziehkurve mit einem maximalen Risiko von 2 Prozent. Der maximale Rückgang in diesem Zeitraum betrug etwa 18 Prozent. Abbildung 4 zeigt das Ergebnis der Erhöhung des maximalen Handelsrisikos auf 6 Prozent. Die Wirkung ist dramatisch: Der Zuwachs ist auf 50% gestiegen. 1/1/94 2/1/95 1/9/96 1/2/97 1/2/98 1/4/99 1/3/00 1/2 / 01 1/2/02 STRATEGIE ZUSAMMENFASSUNG Profitabilität Reingewinn (): 99,997,48 Jahresüberschuss (): 100,00 Exposure (): 29,99 Profitfaktor: 1,50 Auszahlungsquote: 1,64 Recovery-Faktor: 3,21 Drawdown Max. DD (): Die längsten flachen Tage: 19,59 685 Handelsstatistik Nr. Handwerk: 292 Gewinn / Verlust (): 45.21 Durchschnittsalter: Gewinn / Verlust (): 0,73 Durchschnitt. Haltezeit: 37,60; Gewinn (Gewinner): 6,22 Durchschn. Haltezeit (Gewinner): 56.30 Durchschnittl. Verlust (Verlierer): Durchschn. Haltezeit (Verlierer): Max. Konsek. Gewinn / Verlust: -3,80 22,17 5/7 zu 2), wäre der begleitende Drawdown fast unmöglich gewesen, Magen. Exposure kletterte in der Nähe von 70 Prozent, und die längste Wartezeit zwischen neuen Aktienhöhen war mehr als 750 Handelstagen. Der 100-20-Kanal-Ausbruch verlief recht gut bei diesem Test. Wie im Aktienhandelssystemlabor diskutiert, können Sie mit dem System experimentieren, indem Sie die Kanalperioden für jeden Markt optimieren. Zusammengestellt von Dion Kurczek von Wealth-Lab Inc. PERIODISCHE RETURNS Durchschn. Sharpe Bester Worst Prozentsatz Max. Max. Rücklaufverhältnis Rückkehr rentabel konsek. Konsek. Perioden rentabel unrentabel Wöchentlich Monatlich Jährlich 0,15 0,66 8,58 0,64 0,61 0,55 0,58 7,29 -6,14 12,32 -9,25 24.09 -8.25 29.99 -8.91 52.07 55.08 40.00 66.67 6 6 4 3 9 6 6 2 Quartal 1.99 LEGENDE: Reingewinn am Ende der Testperiode, Weniger Provisionen Exposure des Bereichs der Eigenkapitalkurve, der Long - oder Short-Positionen ausgesetzt ist, im Gegensatz zu Cash Profitfaktor Bruttogewinn geteilt durch Bruttoverlust Auszahlungsquote durchschnittlicher Gewinn aus Gewinntrades dividiert durch durchschnittlichen Verlust verlierenden Trades Erholungsfaktor Nettogewinn dividiert durch max. Drawdown Max DD () größter prozentualer Rückgang des Eigenkapitals Längste flache Tage längste Periode, in Tagen ist das System zwischen zwei Aktienhöhen Nr. Trades Anzahl der durch das System erzeugten Trades Gewinne / Verluste () der Prozentsatz der Gewinne, Gewinnt den durchschnittlichen Profit für alle Trades Avg. Haltezeit die durchschnittliche Haltedauer für alle Trades Avg. Gewinn (Gewinner) der durchschnittliche Gewinn für Gewinner Trades Avg. Haltezeit (Sieger) die durchschnittliche Haltezeit für den Siegertrag Avg. Verlust (Verlierer) der durchschnittliche Verlust für verlieren Trades Avg. Haltezeit (Verlierer) die durchschnittliche Haltezeit für verlierende Trades Max. Konsek. Gewinn / Verlust die maximale Anzahl der aufeinanderfolgenden gewinnen und verlieren Trades LEGENDE: Durchschn. Geben Sie den durchschnittlichen Prozentsatz für den Zeitraum Sharpe Ratio durchschnittliche Rendite geteilt durch Standardabweichung der Renditen (annualisiert) besten Rendite besten Rendite für den Zeitraum schlechtesten Rendite schlechtesten Rendite für die Periode profitabel Perioden der Prozentsatz der Perioden, die rentabel waren Max. Konsek. Profitabel die größte Anzahl von aufeinander folgenden profitablen Perioden Max. Konsek. Unrentabel die größte Anzahl von aufeinanderfolgenden unrentablen Perioden Trading System Lab Strategien werden auf Portfolio-Basis getestet (sofern nicht anders angegeben) mit Wealth-Lab Inc. s Testplattform. Wenn Sie ein System haben, das Sie gerne sehen möchten, senden Sie bitte die Handels - und Geldmanagementregeln an editorialactivetradermag. Disclaimer: Das Trading System Lab ist nur für Bildungszwecke gedacht, um eine Perspektive auf verschiedene Marktkonzepte zu liefern. Es ist nicht dazu gedacht, irgendein Handelssystem oder - ansatz zu empfehlen oder zu fördern. Händler werden geraten, ihre eigene Forschung und Prüfung zu tun, um die Gültigkeit einer Handelsidee zu bestimmen. Die bisherige Performance stellt keine Garantie für zukünftige Ergebnisse dar. Historische Tests können kein Systemverhalten im Echtzeit-Handel widerspiegeln. 17 activetradermag Juni 2003 ACTIVE TRADER FUTURES Trading System Lab ABBILDUNG 1 EQUITY CURVE Das System erzielte einen bescheidenen Gewinn, bei dem Long Trades eine Outperformance gegenüber Shorts hatten. 24.000 22.000 20.000 18.000 Kontostand () 60-minütiges Breakout-System Markt: Futures (Indizes). System-Konzept: Dies ist ein Intraday-System, das auf Ausbrüche der Bandbreite in der ersten Stunde des Handels gegründet. Für eine ausführliche Erläuterung der Strategie lesen Sie bitte den Bestand Trading System Lab auf p. 50. Es war beabsichtigt zu sehen, wie das System auf Aktienindex-Futures statt Einzelaktien durchgeführt wurde. In diesem Test wurden die SP 500 (SPY) und Nasdaq 100 (QQQ) Index-Tracking-Aktien als Proxies für die SP 500 und Nasdaq 100 Futures verwendet. Eintragungsregeln: Long Trades: Kaufen, wenn der Schlusskurs der dritten 30-Minuten-Bar über dem hohen der ersten 60 Minuten des Tages liegt. Short-Trades: Short verkaufen, wenn der Schlusskurs der dritten 30-Minuten-Bar unter dem Tiefstand der ersten 60 Minuten des Tages liegt. Beenden: Beenden Sie alle Positionen auf Signale in die entgegengesetzte Richtung oder am Ende des Tages. Geldmanagement: Um das Gewicht der beiden Märkte auszugleichen, werden für jeden Handel 49 Prozent des derzeitigen Portfoliokapitals zugewiesen. Zum Beispiel, wenn das gesamte Eigenkapital bewegt sich bis zu 22.000 und unsere Strategie generiert ein neues Signal, würden wir 10.780 für das nächste Signal zu investieren. Wir verwenden 49 Prozent, um uns etwas Spielraum für Kommission zu geben. Bitte beachten Sie, dass wir das Portfolio-Ergebnis und nicht das einzelne Ergebnis verwenden. Dies ist sehr wichtig und sollte immer verwendet werden, da nur diese Methode spiegelt, was Sie tatsächlich später Erfahrung in Ihrem Trading. Eigenkapital: 20.000 (nominal). Abzug von 0,01 pro Aktie Schlupf und Provisionen. Prüfungszeitraum: Oktober 2001 bis Oktober 2003. 16.000 14.000 12.000 10.000 8.000 6.000 4.000 2.000 0 10/15/01 Eigenkapital 1/8/02 4/1/02 Bargeld 6/24/02 9/23/02 Lange 12/30 / 02 Short 2/2/03 6/26/03 9/25/03 Buy Hold Quelle für alle Zahlen: Wealth-Lab Inc. (Wealth-Lab) ABBILDUNG 2 DRAWDOWN CURVE Der größte Drawdown erfolgte früh im Testzeitraum. Die Systeme der größten Verlust von Trades waren sieben. 0 -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 3/14/02 5/30/02 8/9/02 10/28/02 21/21 / 03 4/2/03 6/13/03 8/29/03 10/15/01 1/2/02 Prüfdaten: SPY und QQQ. Der SPY ist entworfen, um an einem Zehntel des Niveaus des SP 500 zu handeln, der QQQ ist entworfen, um an ein-Fortieth des Nasdaq 100 zu handeln. Wie Futures, die uptick Regel, zum der kurzen Positionen einzugehen, trifft auf diese Instrumente nicht zu. QQQ und SPY können intraday gehandelt werden, haben aber den Vorteil, dass kein Rollover alle drei Monate auftritt. Wir heruntergeladen mehr als zwei Jahre von 30-Minuten-Bars aus der QCharts historischen Intraday-Datenbank für SPY und QQQ. Es gibt ein paar interessante Dinge zu beachten. Für den ersten Stundenbereich nehmen wir die Preise von 9:30 Uhr bis 10:30 Uhr und für die Schlusszeit verwenden wir 4:15 pm Dies ist wichtig, weil wir alle Positionen, die nicht durch ein entgegengesetztes Signal am Ende der Tag. Testergebnisse: Die Ergebnisse für die beiden Jahre sind sehr erfreulich: ein Gewinn von 19,88 Prozent am Startkapital in zwei Jahren gegenüber einem unveränderten Ergebnis für die kombinierten Aktien der beiden Indizes (siehe Abbildung 1). Das System erzielte am 21. Februar 2002 seinen größten Drawdown (-13,52 Prozent) (siehe Abbildung 2). Kauf und hält den größten Drawdown (am 9. Oktober 2002) war -44,87 Prozent. Activetradermag Januar 2004 ACTIVE TRADER 18 ABBILDUNG 3 BEISPIELHANDEL Die durchschnittliche Haltezeit für beide gewinnende und verlierende Trades betrug rund sieben Tage. Auf der Unterseite lag der durchschnittliche Profit pro Handel nur 0,05 Prozent oder 4,46. Dies kann zu wenig sein, um das System wirklich zu handeln. Betrachtet man die Statistiken, ist es interessant festzustellen, dass von den 892 Trades in den letzten zwei Jahren nur 102 durch das entgegengesetzte Signal gestoppt wurden, während der Rest in der ursprünglichen Richtung blieb. Es scheint, dass in den meisten Fällen, sobald der Markt beginnt eine intraday Trend, fährt es in dieser Richtung den ganzen Tag. Abbildung 3 zeigt einen kurzen Handel am 10. September 2003, der am Ende des Tages beendet wurde. Am nächsten Tag erschien der Markt weiter. Wir erhielten ein kurzes Signal, wurden aber gestoppt, nachdem der Markt zurückgekehrt war. Nasdaq 100 Index-Tracking-Aktie (QQQ), 30-Minuten 34.10 34.00 33.90 33.80 Verkauf 33.70 33.60 33.50 Kaufen 33.40 33.30 Verkaufen 33.20 33.10 Kaufen 9/11/03 33.00 Fazit: Es gibt einen großen Unterschied zwischen Indizes und Aktien in Bezug auf diese System. Einzelne Aktien sind tendenziell viel volatiler als ein Index auch, 9/10/03 mit einem Index sehen Sie kaum große Nachtlücken. Dies kann ein Grund dafür sein, dass das 60-minütige Breakout-System viel besser auf den ETFs abläuft als auf den einzelnen Aktien. Wegen des geringen durchschnittlichen Gewinns pro Handel erfordert das System mehr Feinabstimmung. Dennoch macht das Verhältnis der Trades, die ihre ursprüngliche Position für den ganzen Tag gehalten, diese Strategie würdig weiter zu untersuchen. Volker Knapp von Wealth-Lab Inc. STRATEGIE ZUSAMMENFASSUNG Profitabilität Nettogewinn (): Nettogewinn (): Exposure (): Profitfaktor: Payoff-Verhältnis: Recovery-Faktor: Drawdown Max. DD (): Längste flache Tage: 3.976,80 19,88 44,95 1,11 0,93 1,38 -13,52 1,742 Handelsstatistik Anzahl der Titel: Gewinn / Verlust (): Durchschn. Gewinn / Verlust (): Durchschn. Haltezeit: Durchschn. Gewinn (Gewinner): Durchschn. Haltezeit (Gewinner): Durchschn. Verlust (Verlierer): Durchschn. Haltezeit (Verlierer): Max. Konsek. Sieg / Verlust: 892 54.60 0.05 7.21 0.80 7.35 -0.86 7.04 10/7 PERIODISCHE RÜCKGABE Durchschn. Sharpe Bester Worst Prozentsatz Max. Max. Rücklaufverhältnis Rückkehr rentabel konsek. Konsek. periods profitable unprofitable Weekly Monthly 0.19 0.77 0.80 0.88 1.39 7.31 -3.36 6.97 -3.17 6.19 -2.19 53.85 52.00 66.67 7 2 6 8 4 2 Quarterly 2.07 LEGEND: Net profit Profit at end of test period, less commission Exposure The area of the equity curve exposed to long or short positions, as opposed to cash Profit factor Gross profit divided by gross loss Payoff ratio Average profit of winning trades divided by average loss of losing trades Recovery factor Net profit divided by max. drawdown Max. DD () Largest percentage decline in equity Longest flat days Longest period, in days, the system is between two equity highs No. trades Number of trades generated by the system Win/Loss () The percentage of trades that were profitable Avg. gain The average profit for all trades Avg. hold time The average holding period for all trades Avg. gain (winners) The average profit for winning trades Avg. hold time (winners) The average holding time for winning trades Avg. loss (losers) The average loss for losing trades Avg. hold time (losers) The average holding time for losing trades Max. consec. win/loss The maximum number of consecutive winning and losing trades LEGEND: Avg. return The average percentage for the period Sharpe ratio Average return divided by standard deviation of returns (annualized) Best return Best return for the period Worst return Worst return for the period Percentage profitable periods The percentage of periods that were profitable Max. consec. profitable The largest number of consecutive profitable periods Max. consec. unprofitable The largest number of consecutive unprofitable periods Trading System Lab strategies are tested on a portfolio basis (unless otherwise noted) using Wealth-Lab Inc. s testing platform. If you have a system youd like to see tested, please send the trading and money-management rules to editorialactivetradermag. Disclaimer: The Trading System Lab is intended for educational purposes only to provide a perspective on different market concepts. It is not meant to recommend or promote any trading system or approach. Traders are advised to do their own research and testing to determine the validity of a trading idea. Past performance does not guarantee future results historical testing may not reflect a systems behavior in real-time trading. ACTIVE TRADER January 2004 activetradermag 19 Four-percent breakout system Market: Nasdaq 100 index-tracking stock (QQQ). System concept: This system is from the book Trade like a Hedge Fund (John Wiley Sons, 2004) by James Altucher. Both day traders and hedge-fund managers love to fade (i. e. trade in the opposite direction of) sharp intraday moves. The thought behind this type of trading is price is unlikely to go much higher after an extreme up move. Traders take short positions anticipating a reversal. In some situations, however, the market ignores the contrarians and continues to rise. Traders who fade the up move must cover their short positions, which leads to panic buying and further upward momentum. The four-percent breakout system is an attempt to quantify and profit from this market scenario. The system goes long when price rises four percent from the previous close in this case, assumed to be the point at which short sellers must concede they were wrong and cover their positions, driving prices even higher during the trading day. The system is long only no short trades are made. Rules: Entry Buy today if price gains four percent from the previous trading days closing price. Exit Exit on the open of the next trading day. Figure 1 shows sample trades in QQQ from March and April 2003. Risk control and money management: This system tests only one market, and enters only one position at a time, so 100 percent of equity should be tied up on each trade. Starting equity: 100,000. Deduct 20 per round-turn trade for slippage and commissions. Test data: The system was initially tested only on the QQQ. It was also tested on the Active Trader Standard Stock Portfolio, which contains the following 18 stocks: Apple Computers (AAPL), Boeing (BA), Citigroup (C), Caterpillar (CAT), Cisco Systems (CSCO), Disney (DIS), General Motors (GM), HewlettPackard (HPQ), International Business Machines (IBM), Intel (INTC), International Paper (IP), J. P. Morgan Chase (JPM), Coca-Cola (KO), Microsoft (MSFT), Sears (S), Starbucks (SBUX), ATT (T) and Wal-Mart (WMT). Test period: March 1999 through June 2004 for the QQQ test July 1994 to June 2004 for the Active Trader portfolio. 20 FIGURE 1 SAMPLE TRADES March and April 2003 were very active months for the four-percent breakout system. There was one large winner, two mid-size winners and one large losing trade. Nasdaq 100 index-tracking stock (QQQ), daily Sell Sell Sell Sell Buy Buy 27.40 27.20 27.00 26.80 26.60 26.40 26.20 26.00 25.80 25.60 25.40 25.20 25.00 24.80 24.60 24.40 24.20 24.00 23.80 23.60 23.40 100M 50M April 2003 Source for all figures: Wealth-Lab Inc. (wealth-lab) Buy Buy Volume FIGURE 2 EQUITY CURVE The equity grew steadily from 2000 through 2002, but the system has been stagnating since mid-2002. 250,000 240,000 230,000 220,000 210,000 200,000 190,000 180,000 170,000 160,000 150,000 140,000 130,000 120,000 110,000 100,000 90,000 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 3/10/99 9/1/99 3/1/00 9/1/00 3/1/01 9/4/01 3/8/02 9/5/02 3/6/03 9/3/03 3/3/04 Equity Cash Account balance () activetradermag September 2004 ACTIVE TRADER FIGURE 3 DRAWDOWN CURVE The drawdown phase from mid-2002 to the present dominates the drawdown curve. 0 -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 -11 3/10/99 9/1/99 3/2/00 9/1/00 3/5/01 9/4/01 3/8/02 9/6/02 3/7/03 9/5/03 3/5/04 Test results: Figure 1 shows the first trade was a success. Price rose 40 cents after entry and the market made a small gap open the following day for a twopercent profit. The next two trades, which occurred only a few days later, were not as successful but nonetheless booked modest profit. However, the next trade wiped out the previous profit and then some. Price gapped up at the market open, beyond the four-percent threshold, and the entry order was filled (this particular trade would have probably been subject to negative slippage because of the volatility at the open). Price then suddenly reversed, and the result was a large loss upon the exit the following day. The fact that the initial losing trade occurred on a day when prices gapped above the entry level on the open suggests the system might benefit from a filter that ignores the signal if price opens with a greater than four-percent gain. The equity curve (Figure 2) provides a better indication of the systems overall performance. After a small loss in 1999, profits began in early 2000 and lasted until mid - to late 2002. The drawdown curve (Figure 3) confirms this, as the 12-percent drawdown began in late 2002. The system is more or less flat from April 2003 forward. The only trade after that was in July 2003, resulting in a loss of 0.08 percent. Drawdown the designers knew of previous QQQ price movement), it was tested on other markets in an attempt to determine its validity. Our starting equity for the Active Trader portfolio test was also 100,000, although only 10 percent of equity was committed per trade. This equity curve (Figure 4, p. 60) shows fairly steady growth from the beginning of the test period through mid2002. From that point, there is a slight decline in capital and a general stagnation as fewer trades take place. This equity curve mirrors the QQQ equity curve. However, the fact that the system was profitable on a portfolio of stocks (8.95 percent annualized gain) and not just one stock is evidence the system is based on a valid core assumption. System variation: James Altucher publishes a variation of the system that adds one additional entry rule: Price must be down two percent on the day before entering a trade. This rule is intended to avoid entering when a price move is nearly exhausted, and allows the system to capture solid rebound PERIODIC RETURNS Portfolio test results: While it is still too early to tell if this system is worth trading on the QQQ (because its possible the system was subconsciously designed to take advantage of what STRATEGY SUMMARY Profitability Net profit (): Net profit (): Exposure (): Profit factor: Payoff ratio: Recovery factor: Drawdown (): Max. DD (): Longest flat days: 121,023 121.09 7.81 1.82 1.13 4.05 Trade statistics No. trades: Win/loss (): Avg. trade (): Avg. winner (): Avg. loser (): Avg. hold time: Avg. hold time (winners): Avg. hold time (losers): Max. consec. win/loss: 105 64.76 0.80 2.38 -7.55 1.00 1.00 1.00 11/5 Avg. Sharpe Best return ratio return Weekly Monthly 0.30 1.30 1.18 1.35 1.03 0.56 11.89 13.34 25.74 75.51 Worst Percentage Max. Max. return profitable consec. consec. periods profitable unprofitable -9.23 22.10 4 63 -6.40 -6.98 -2.52 50.00 59.09 66.67 10 8 4 15 5 2 Quarterly 3.93 Annually 16.71 29,900 -11.93 420 The system remains flat much of the time. A flat period is considered unprofitable for purposes of this report. LEGEND: Net profit Profit at end of test period, less commission Exposure The area of the equity curve exposed to long or short positions, as opposed to cash Profit factor Gross profit divided by gross loss Payoff ratio Average profit of winning trades divided by average loss of losing trades Recovery factor Net profit divided by max. drawdown Max. DD () Largest percentage decline in equity Longest flat days Longest period, in days, the system is between two equity highs No. trades Number of trades generated by the system Win/Loss () the percentage of trades that were profitable Avg. trade The average profit/loss for all trades Avg. winner The average profit for winning trades Avg. loser The average loss for losing trades Avg. hold time The average holding period for all trades Avg. hold time (winners) The average holding time for winning trades Avg. hold time (losers) The average holding time for losing trades Max. consec. win/loss The maximum number of consecutive winning and losing trades LEGEND: Avg. return The average percentage for the period Sharpe ratio Average return divided by standard deviation of returns (annualized) Best return Best return for the period Worst return Worst return for the period Percentage profitable periods The percentage of periods that were profitable Max. consec. profitable The largest number of consecutive profitable periods Max. consec. unprofitable The largest number of consecutive unprofitable periods Trading System Lab strategies are tested on a portfolio basis (unless otherwise noted) using Wealth-Lab Inc. s testing platform. If you have a system youd like to see tested, please send the trading and money-management rules to editorialactivetradermag. Disclaimer: The Trading System Lab is intended for educational purposes only to provide a perspective on different market concepts. It is not meant to recommend or promote any trading system or approach. Traders are advised to do their own research and testing to determine the validity of a trading idea. Past performance does not guarantee future results historical testing may not reflect a systems behavior in real-time trading. ACTIVE TRADER September 2004 activetradermag 21 Account balance () moves. This generally increases the efficiency of the system while reducing the number of actual trades. The bottom equity curve in Figure 4 shows the results of the system variation on the Active Trader portfolio. Since the new filter reduces the number of trades, the position size was changed to 25 percent for each trade. The shape of the equity curve is similar to the previous run, but actual profit is higher and the system does not enter as many losing trades during the stagnation period of mid2002 to present. Bottom line: The four-percent breakout system could not be much simpler. Simpler systems are often the most effective, and this one is no exception. However, there needs to be sufficient post-publication data to provide a reliable test for the QQQs. This system is from James Altuchers Trade like a Hedge Fund. Compiled by Volker Knapp of Wealth-Lab FIGURE 4 EQUITY CURVE: ACTIVE TRADER PORTFOLIO The upper equity curve shows the results of the four-percent breakout system on the Active Trader standard stock portfolio using 10 percent of equity per trade. The lower equity curve is a system variation that enters after a down move and uses 25 percent of equity per trade. 200,000 150,000 100,000 50,000 0 250,000 200,000 150,000 100,000 50,000 0 7/15/94 7/31/95 6/7/96 6/2/97 6/1/98 5/6/99 5/1/00 5/1/01 5/1/02 5/1/03 4/5/04 Equity Cash A B 22 activetradermag September 2004 ACTIVE TRADER TRADING Strategies BROADENING PATTERNS: Clues to breakout direction A partial rise or decline can predict the direction of a breakout. Learn to use these signals to increase profits when trading broadening patterns. BY THOMAS N. BULKOWSKI T rying to determine when a breakout will occur in broadening chart patterns, which are expanding rather than contracting price formations, can be difficult. However, partial rises (PRs) or FIGURE 1 PARTIAL RISE partial declines (PDs) can improve the odds of making a correct decision. These signals predict immediate breakouts and indicate their direction, too, allowing you to increase your profits and reduce your losses. However, because a PR or PD often slows overall momentum, the size of the eventual breakout is not as large as when a PR or PD does not appear. Broadening tops and bottoms Figure 1 shows two broadening bottom patterns. These are different from broadening tops because price enters the pattern from the top. In both patterns, price touches each trendline at least two times and swings in a progressively wider range. That is, the minor highs get higher and the minor lows get lower. The July pattern shows a PR, which occurs after the pattern is established that is, there were at least two touches of each trendline before the PR. Price makes a partial rise when it leaves the bottom trendline and works its way higher but fails to touch or come too close to the top trendline before turning away. How close is close Use the figures in this article and your common sense as guides. For example, the July broadening bottom has three top trendline touch - The July broadening bottom pattern appeared midway through the down move. A partial rise accurately signaled a downside breakout from the pattern. Milacron Inc. (MZ), daily 34 32 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 1998 Mar. Apr. May June July Source: Proprietary software (Thomas Bulkowski) Aug. Sept. Oct. Nov. Dec. Partial rise 12 34 ACTIVE TRADER April 2004 activetradermag 23 even better 80 percent (see Table 1, top right, for more statistics). The percentages reflect how often partial rises and partial declines predicted Notice how the July pattern is midbreakout direction. way between the price at the start of the downtrend (around 32) and its low Chart pattern Partial decline Partial rise (around 14). The middle of the pattern is Broadening bottom 80 67 around 23, the center of the 32-14 range. Although broadening patterns someBroadening top 65 86 times act as half-staff patterns that Broadening wedge, ascending Not measured 84 form in the middle of moves, broadenBroadening wedge, descending 76 Not measured ing bottoms usually function as reversals in a downtrend, not as continuation patRight-angled and ascending Not measured Not measured terns within those trends, as they do in Right-angled and descending 78 58 Figure 1. Figure 2 includes two broadening tops with PDs. In a partial decline, price es, not two or four: The second minor 500 stocks from mid-1991 to mid-1996, a leaves the top trendline and descends high (point 2) comes close enough to call bull market, showed that a PR correctly but does not come close to or touch the it a touch, but the third (point 3) does predicted a downward breakout 67 per - bottom trendline. An upward breakout cent of the time. The accuracy rate of usually follows immediately. Again, the not. Analysis of 77 broadening bottoms on PDs predicting upside breakouts was broadening top pattern must touch each trendline at least two times before a PD signal can occur. FIGURE 2 BROADENING TOPS Table 1 shows PDs in broadening tops correctly Both of these broadening tops included partial declines, which predicts an upward breakpredicted an upward breakout of the pattern. out 65 percent of the time, Newport Corporation (NEWP), daily while partial rises were 86percent accurate in predict7 ing downside breakouts. In a larger combined study of broadening tops 6 and bottoms, PDs worked 77 percent of the time. When a 5 PD occurred, the post-breakPartial decline out up move was 32 percent without a PD, the rise measured 36 percent. Thus, the 4 PD affected momentum by reducing the eventual rally. PDs not resulting in breakouts occurred just nine perPartial decline cent of the time, which 3 means false signals are comparatively rare. For PRs, the post-breakout decline measured 15 percent without a PR, the declines averaged 17 percent, indicat2 ing a partial rise steals ener1997 June July Aug. Sept. Oct. Nov. Dec. 1998 Feb. Mar. Apr. May June gy from the resulting down Source: Proprietary software (Thomas Bulkowski) move. 24 activetradermag April 2004 ACTIVE TRADER TABLE 1 PARTIAL RISES AND DECLINES: SUCCESS RATES FIGURE 3 TRENDLINE TOUCHES Look for a partial rise or decline only after price touches each trendline of the broadening pattern at least twice. Here, a partial rise formed in this ascending, right-angled broadening formation. Tommy Hilfiger (TOM), daily Partial rise 37 35 33 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 1998 Feb. Mar. Apr. May Source: Proprietary software (Thomas Bulkowski) June July Aug. Sept. Oct. False breakout signals for PRs (i. e. when a partial rise occurred inside a broadening top pattern after price touched each trendline twice without triggering a breakout) occurred just 11 percent of the time in the 350 patterns examined. A broadening top usually acts as a continuation pattern within the prevailing price trend, as shown in Figure 2. 1 2 Not a partial decline 3 Right-angled broadening formations Figure 3 shows a rightangled, ascending broadening formation. The top trendline slopes upward (ascends) and the bottom trendline is horizontal or nearly so. Like other broadening patterns, the breakout can occur in any direction, but this pattern usually reverses the trend. The figure shows this, as prices rise into the pattern and exit out the bottom. After two touches of each trendline occur, look for a partial rise or decline. The late-May decline in Figure 3 does not show a partial decline. Why Because the pattern at that point did not have at least two minor touches of each trendline. Price touches at point 1 but it is not a minor high or low, so it does not count as a touch. Point 2 is valid, as is point 3. Only after price touches point 3 can you draw the horizontal trendline. By that time, the three touches on the top connect an up-sloping trendline. The partial rise that follows correctly predicts a downward breakout. Figure 4 shows a descending right-angled broadening FIGURE 4 REVERSAL PATTERN Although broadening formations are often continuation patterns, descending, right-angled broadening formations like the one shown here usually act as reversal patterns. CDI Corp. (CDI), daily Partial rise 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 Sept. 1994 Nov. Dec. 1995 Feb. Mar. Source: Proprietary software (Thomas Bulkowski) Apr. May June July Aug. ACTIVE TRADER April 2004 activetradermag 25 FIGURE 5 DESCENDING BROADENING WEDGE The August pattern did not produce a valid partial decline because price must drop from the top trendline and curl around. Here, price rose from the bottom trendline. The first partial decline in the October pattern fails to predict an immediate upward breakout, but is correct in the longer term. Transocean Inc. (RIG), daily 52 50 48 46 44 42 40 38 36 34 32 31 30 29 28 27 26 25 24 23 22 21 Dec. 2000 Feb. Mar. 20 Not a partial decline pattern. The top trendline is horizontal and the bottom one slopes down. Price touches the bottom trendline, bounces up but does not come close to or touch the top trendline before retracing its gains. This PR predicted a downward breakout. As is the case with this example, the descending, right-angled broadening pattern usually acts as a price reversal. In the descending pattern, partial rises worked just 58 percent of the time and partial declines worked 78 percent of the time in descending, right-angled broadening patterns. Broadening wedges Figure 5 shows a descending broadening wedge, which consists of two down-sloping trendlines (think of a downward-tilting megaphone). The rules for wedges are the same as other broadening patterns: There must be at least two minor high touches of the top trendline and at least two minor low touches of the bottom trendline. Only then is the pattern valid and only then should you look for a partial rise or decline. The pattern usually acts as a continuation, rather than a reversal, of the prevailing price trend. However, the two wedges shown in Figure 5 are reversal patterns. In the August pattern, prices climbed into the pattern and broke out to the upside, but the overall trend (except for a few days after the breakout) was downward after the pattern. Prices in the October wedge were trending downward into the pattern and exited out its top. The trend after the pattern ends is predominantly upward. In the August pattern example, the slight dip in Partial declines 1999 June July Aug. Sept. Oct. Source: Proprietary software (Thomas Bulkowski) Nov. FIGURE 6 ASCENDING BROADENING WEDGE After the pattern is established, a partial decline fails to correctly predict an upward breakout. Later, a partial rise precedes a downside breakout. WPS Resources Corp. (WPS), daily 32 31 30 Partial rise 29 28 27 26 25 24 23 Failed partial decline 22 21 20 June 1999 Aug. Sept. Oct. Nov. Dec. Source: Proprietary software (Thomas Bulkowski) 2002 Feb. Mar. Apr. May 26 activetradermag April 2004 ACTIVE TRADER early September was not a partial decline. Price in a PD must start from the top trendline, bow downward (without coming close to or touching the bottom trendline) and rejoin the top trendline. In this case, price leaves the bottom trendline, not the top one. In the October pattern, the first partial decline is a failure because price does not breakout upward immediately after touching the top trendline. Instead, price drops down again and finally shoots out the top. A partial decline correctly predicts an upward breakout 76 percent of the time. Not enough samples were found for partial rises in descending broadening wedges. Figure 6 (p. 28) shows an ascending broadening wedge. Both trendlines slope upward and minor highs and minor lows touch each trendline at least twice. The January partial decline failed because price did not break out to the upside it touched the top trendline, then reversed. The partial rise does better when it leaves the bottom trendline, bounces up and then plunges through the bottom trendline. A PR correctly predicts a downward breakout 84 percent of the time. Additional research Books by Thomas Bulkowski: Encyclopedia of Chart Patterns (John Wiley Sons, 2000) Trading Classic Chart Patterns (John Wiley Sons, 2002) Active Trader articles: Technicals meet fundamentals in the earnings flag, February 2004, p. 30 A different breed of scallop, January 2004, p. 32 The three rising valleys pattern, December 2003, p. 28 Pipe bottom reversals, November 2003, p. 28 Grabbing the bull by the horns, September 2003, p. 46 Head-and-shoulders bottoms: More than meets the eye, August 2003, p. 32 The high-low game, July 2003, p. 28 Tom Bulkowskis scientific approach, September 2002, p. 32 ACTIVE TRADER April 2004 activetradermag 27 TRADING Strategies HIGH, TIGHT FLAG helps squeeze out profits This bullish formation boasts excellent post-breakout performance and a low failure rate exactly the type of pattern traders should look for in bull markets. BY THOMAS N. BULKOWSKI A high, tight flag (HTF) is a consolidation pattern that forms after a stocks price doubles. When price breaks out above the pattern, it signals the rise is not over. Figure 1 shows an example of an HTF that formed in January-February 2000. The uptrend started in October at a low of 5.50 and reached a high of 11.35 at the HTFs starting point a doubling of price in less than a month. Although many HTFs have irregular shapes, you can usually draw a trendline along the top of the pattern to signal a breakout. In this example, parallel trendlines mark the flags upper and lower boundaries. Volume slopes downward over the course of the flag, as it did in 90 percent of HTFs in a recent study. The basic HTF trade strategy is to buy at the close of the day after price breaks out above the patterns upper trendline. In Figure 1, the stock rallied 52 percent from the closing price the day after price pierced the HTFs upper trendline to the ultimate high. Although it is sometimes difficult to buy high and sell higher, the price moves following HTFs show how such an approach can work. Flag criteria What should you look for when selecting HTFs That depends on whom you ask. William ONeil, who popularized the pattern, has several selection criteria (see Additional reading, p. 33). He has written the rally preceding the pattern should measure 100 to 120 percent and take less than two months the flag should move sideways for three to five weeks. Finally, the flag should retrace no more than 20 percent of the preceding rally. Applying these rules to 252 patterns found in price data of approximately 500 stocks between mid-1991 and early 2004 filtered out all of them (An earlier study found only six of 81 patterns met his criteria these patterns did, however, produce average gains of 69 percent.) The flag shown in Figure 1 actually does not meet ONeils criteria because it retraced 52 percent of the prior rise (most flags failed ONeils filter because they retraced more than 20 percent) and the flag duration lasted more than seven weeks. 28 activetradermag December 2004 ACTIVE TRADER FIGURE 1 A LARGE HIGH, TIGHT FLAG This pattern does not meet William ONeils HTF criteria, but the post-breakout rise was 52 percent. Such a well-defined flag shape is unusual. Alkermes (ALKS), daily Ultimate high 17 16 15 14 13 12 11 10 High, tight flag 9 8 7 6 Trend start 5 Another study that used different selection criteria for HTFs also showed an average post-breakout gain of 69 percent. The criteria for this study was simply a near doubling (a rise of 90 percent or more in less than two months) of the stock price, which is easy to find by looking for stocks that have moved up sharply in less than two months, then searching for a nearby consolidation region. The study placed no limit on flag length, although HTFs equal to or shorter than the 14-day median length performed better (71 percent) than those that were longer (66 percent). The study also ignored the size of the retracement. 1998 Aug. Sept. Oct. Nov. Dec. 1999 Feb. Mar. Apr. May June 4 Source: Proprietary software (Thomas Bulkowski) FIGURE 2 TWO HTFS The first flag looks like a descending, broadening wedge and the second like a falling wedge. Both HTFs show good gains after the breakout with the first pattern hitting overhead resistance at the ultimate high. National Semiconductor Corp. (NSM), daily Ultimate high 84 76 68 62 56 50 44 40 36 32 28 24 22 20 18 16 14 12 10 Trend start 8 6 1999 May June July Aug. Sept. Oct. Source: Proprietary software (Thomas Bulkowski) Nov. Dec. 2000 Feb. Mar. HTF examples Figure 2 shows two HTFs identified in the study. The trend start point was determined by finding a 20-percent reversal of the existing trend, measured from a prior low to the most recent close. The ultimate high was identified by finding a subsequent 20-percent trend change, measured from a prior high to the recent close. HTF 1 was preceded by a 156-percent rally that lasted 40 days. The flag retraced 38 percent of the rally and lasted 15 days. After the breakout, price climbed 54 percent to the ultimate high, which occurred at a resistance area established by price peaks as far back as mid-1995 (not shown). Price rose 121 percent leading up to HTF 2, taking 51 days to make the climb. HTF 2 Ultimate high Trend start HTF 1 ACTIVE TRADER December 2004 activetradermag 29 FIGURE 3 TWO MORE HTFS The first HTF launches from a flat base and price soars to the ultimate high. The second trade is more typical with the rise to the ultimate high about half the distance, on a percentage basis, from the trend start to the flag. Vertex Pharmaceuticals (VRTX), daily 93 85 77 71 65 59 53 49 45 41 37 33 29 25 23 21 19 17 15 13 Trend start 11 May June July Aug. Sept. 9 Ultimate high Ultimate high HTF 2 HTF 1 Trend start 1999 Dec. 2000 Feb. Mar. Apr. Source: Proprietary software (Thomas Bulkowski) FIGURE 4 HTF FAILURE This HTF fails to travel far due to overhead resistance and a change in company fundamentals. The stock tumbles on an earnings warning. Noven Pharmaceuticals (NOVN), daily Resistance Ultimate high 44 42 40 38 36 34 32 30 Dead-cat bounce 28 26 24 23 22 21 20 19 18 17 16 15 14 13 2000 Feb. Mar. Apr. May June July Aug. Sept. Oct. 12 Nov. The flag retraced 45 percent of the rally and was 29 days long. After the breakout, price rallied 92 percent. This pattern did not have overhead resistance to overcome on its way to the ultimate high. That may explain why price nearly doubled before trending downward. Notice the irregular shapes of these two HTFs. The first looks like a small broadening wedge and the second looks like a regular falling wedge. Volume trends downward in both. Figure 3 shows two more examples. Starting from a flat base in late 1999, price climbs 116 percent leading to HTF 1 and soars 129 percent afterward. The stock did not perform as well after HTF 2 because the market changed from bull to bear between the two patterns (the bear market started in March 2000, near HTF 1s ultimate high). The second HTF pattern was preceded by a 136-percent price rise and followed by a 61-percent rally after the breakout. HTF The measure rule The second pattern in Figure 3 is typical of the rise you can expect after an HTF in a bull market. For all 252 patterns in the study, the climb leading to the pattern averaged 124 percent, but the post-breakout gain was just 69 percent. To determine an approximate target, compute the percentage change from the low of the trend start point to the high at the top of the flag. After the breakout, the move from the flags lowest low should measure approximately half this Trend start Source: Proprietary software (Thomas Bulkowski) 30 activetradermag December 2004 ACTIVE TRADER An HTF triggers a buy signal after a stock has made a significant up move and, thus, will appear overbought to many traders. amount. This measure rule works 90 percent of the time in a bull market. close above the highest high in the pattern as the buy signal. This is important: If you buy before the breakout, price might drop instead. Only 10 percent of the 252 patterns in the study failed to climb at least 20 percent, and none failed to climb less than 5 percent. Those are very low failure rates. For protection, use progressive stops. For example, once price makes a new Pattern failures Figure 4 illustrates two types of pattern failures. The first is a rise blocked by overhead resistance. Underlying support or overhead resistance (look for a solid mass of horizontal price movement or peaks and valleys stopping near the same price area) spells death to most chart-pattern breakout trades. In this example, price climbed 31 percent (the pre-pattern rally was 128 percent) after the HTF breakout. That is a significant rally by most standards, but it fails to come close to the 64-percent gain using the measure rule. The second failure comes from the fundamentals. The company issued an earnings warning for the quarter and said fullyear earnings would suffer as well. The stock tumbled 43 percent in one session. Price bounced up during the next month before rounding over and making a lower low in a classic dead-cat bounce (DCB) pattern. Three months later, the stock dropped another 32 percent on a warning about flat annual revenues. Problems cannot always be fixed in one quarter. Avoid a stock showing a DCB for at least six months preferably a year. That will give the company time to get its act together. high, raise the stop to just below the prior minor low, provided it is not too far away otherwise use 1.5 times the daily volatility, which is 1.5 times the average intraday trading range over the prior month. Keep raising the stop as price climbs. Use the measure rule to find a target price (half the price rise leading to the HTF, projected upward from the flag low). Most of the gains (35 percent, on average) will come in the first week, so enter as soon as you get the signal. Also, the time from the trend start to the flag start will be slightly less (by six days on average) than the time from the flags end to the ultimate high. Additional reading Books: How to Make Money in Stocks (McGraw-Hill, 1988) by William ONeil Books by Thomas Bulkowski: Encyclopedia of Chart Patterns (John Wiley Sons, 2000) Trading Classic Chart Patterns (John Wiley Sons, 2002) Active Trader articles: Trading busted patterns, November 2004, p. 42 Half-staff patterns: Profiting from flags and pennants, September 2004, p. 48 Three falling peaks: Bearish trend change pattern, August 2004, p. 32 Chart patterns: Does size matter, June 2004, p. 44 Trading disaster: the dead-cat bounce, May 2004, p. 44 Broadening patterns: Clues to breakout direction, April 2004, p. 36 Technicals meet fundamentals in the earnings flag, February 2004, p. 30 A different breed of scallop, January 2004, p. 32 The three rising valleys pattern, December 2003, p. 28 Pipe bottom reversals, November 2003, p. 28 Grabbing the bull by the horns, September 2003, p. 46 Head-and-shoulders bottoms: More than meets the eye, August 2003, p. 32 Tom Bulkowskis scientific approach, September 2002, p. 32 You can purchase past articles at activetradermag/purchasearticles. htm and download them to your computer. Trading the pattern An HTF triggers a buy signal after a stock has made a significant up move and, thus, will appear overbought to many traders. This is a momentum play, buying high and selling higher. Trading an HTF is like standing on the edge of the cliff and jumping off, hoping the water at the bottom is deep enough. You need a good dose of courage to take the plunge. To trade the pattern, wait for price to either close above the flag trendline or, if the HTF has an irregular shape, use a ACTIVE TRADER December 2004 activetradermag 31 TRADING Strategies Mastering TWO-MINUTE breakouts How can you find consistent trade opportunities One way is to trade breakouts through yesterdays high and low but only after the stock has shown its true colors. trade is to buy upside or downside breakouts of the previous days high or low, respectively, avoiding trades in the middle of the days range. Well show how to apply this technique using twominute charts. The tools For this approach, use a two-minute candlestick chart encompassing a two-day time horizon (today and yesterday). For a long trade, buy the stock once it has cleared the whole number closest to the previous days high. For example, assume a stock made a high of 47.9 yesterday. In this case, you would enter a buy order when the stock hits 48.5 (having cleared 48, the nearest whole number) and when time and sales shows that most trades are being executed at the ask price, which would suggest strong demand for the stock. Reverse the logic for short trades. The reason for placing the entry a certain amount above the previous days high in this case, 48.5 is to make sure the trade safely clears the hurdle of the previous days trading range, accounting for any market noise that may be present. We dont want to buy a double top, we want to buy a breakout above the previous days high. Entering 0.3 to 0.5 above the whole number helps avoid false breakouts. This approach works because many professional traders and institutional buyers buy such breakouts. In addition, some institutional buy programs also factor in the open, high, low and closing prices. When such programs trigger buy signals and money starts flowing into a BY KEN CALHOUN t is often a struggle to find the most appropriate indicator for a given trading situation. A tool that works in one environment may not be appropriate in another. Momentum oscillators or the Nasdaq and SP 500 futures may provide early signals of shifts in the stock market, but these tools also are often unreliable. Moving average crossovers provide trend confirmation but generally lag price action, and you cannot count on sustained trends in consolidating markets. Further, market makers frequently disguise their intentions via Electronic Communications Networks (ECNs) or Level II head fakes, which render the Level II screen more or less useless. So, whats a trader to do Watch price action. Trading breakouts and breakdowns of chart patterns is a reliable and simple trading technique that can help you limit risk. A relatively consistent short-term, pattern-based 32 I stock, you can ride the coattails of the large money on the way up. To make sure, however, dont enter the trade until the stock also has cleared the required noise level. We also use the time and sales window to confirm that any large block trades are going our way and that most trades are executed at the ask price for long trades, or the bid price for short sales. It also is good if a directional chart pattern i. e. one that implies a move either up or down confirms the breakout. A simple example is successive closes at the high (or low) of the price bars leading up to, or coinciding with, the breakout. Also, many traders use specific candlestick chart patterns to indicate likely price direction. The rules The best time to use this method is the profitable and volatile 9:40 a. m. to 11 a. m. (EST) time period. Trades typically last several to 20 minutes. Here are stepby-step guidelines for applying this Strategy snapshot Strategy: Two-day breakout Market: Stocks Entry: Go long (short) on move .3 to .5-points above (below) whole number closest to previous days high (low). Exit: Exit with trailing stop or on close. Risk control: Stop-loss of no more than 0.4 points. Trail stop at this interval if market moves in direction of trade. activetradermag September 2001 ACTIVE TRADER technique. 1. Define the days breakout and breakdown entry levels before each market open. Set up one of your trading screens to plot a single, large two-minute candlestick chart covering two days (today and the previous trading day) of trading activity, as shown in Figure 1. Make sure you start charting by 8:30 a. m. so you can spot any pre-market top or bottom formations, price gaps and trends. Identify the previous days high and low. 2. Enter 0.3 to 0.5 points above the previous days high (for long trades) or low (for short trades). 3. All intraday trades should have a maximum stop-loss of 0.4 points. Combined with entering 0.3 to 0.5 points above the previous days high, this provides an excellent risk management tool. In effect, we will exit if the reason for the trade is negated, i. e. the stock moves FIGURE 1 BUY SIGNAL back into the trading range near the whole number and the entry price level is violated. 4. Trail the stop to protect profits. 5. Because the market often reverses around 10 a. m. each day, it is useful to tighten the stop during this time to three or four spreads (the colored bands of bid and ask levels on the Level II screen) behind the current inside bid. With decimal trading, this allows active traders to keep even tighter stops than was previously possible. Glossary Time and sales: The real-time, official record of executed trades (as opposed to bids and offers) throughout the day. Most trading platforms include a time and sales window to monitor this activity. Noise: Random, meaningless price fluctuations that can knock traders out of the market. Buy programs (program trading): Computer-based trading approach whereby institutions or large trading operations execute large volume in related markets to take advantage of discrepancies between them (i. e. buying SP stocks and selling SP futures). See Program trading and fair value, Active Trader, Jan./Feb. 2001, p. 28, for more information. Uptick rule: Securities and Exchange Commission rule that requires short sales to be executed when the last recorded price in a stock is higher than (or equal to, depending on the circumstances) the immediately preceding price. (The rule varies slightly for NYSE and Nasdaq stocks, although the principle is the same.) See A walk on the short side, Active Trader, July 2000, p. 32, for more information. Trade examples Figure 1 shows that on April 27, Ebay (EBAY) made a high of 48 and a low of 45.5. Based on the guideline to place the entry points 0.3 to 0.5 points above or below the previous days high and low prices, on April 30 we set long entry at 48.5 (0.5 points above the previous days high of 48, which was a whole number). A buy signal occurs in EBAY when the stock moves .5 points above the whole number nearest to yesterdays high. EBay Corp. (EBAY), two-minute 52.00 Buy signal is generated when price exceeds previous days high .5 points. 51.50 51.00 50.50 0 50.00 49.50 Previous days high: 48.00 49.00 48.50 48.00 47.50 47.00 46.50 46.00 45.50 Previous day (compressed) 4/27/01 9:30 4/30/01 10:00 Current day 10:30 11:00 11:30 12:00 45.00 Source: Data Broadcasting Corp. ACTIVE TRADER September 2001 activetradermag 33 We trailed a stop no more than 0.4 points behind the current price level. In this trade, the trailing stop was triggered at 49.375, yielding a net profit of 0.875 points in less than 20 minutes. Figure 2 (left) is an example on the short side of the market. Adobe (ADBE) traded between 42.4 and 45.4 on May 2. The next day (May 3) we therefore looked to go short if the market fell to 41.6, 0.4 points below the whole number (42) closest to the previous days low. However, ADBE gapped down to 41.6 in pre-market trading. When this happens, its a good idea to move the initial entry point farther away from the price action to avoid being caught on the FIGURE 2 SELL SIGNAL wrong side when the market opens. Therefore, we adjusted the entry to 41.4 to clear the gap with as small a distance as possible. This is not an exact science. Sometimes you will jump into a trade too soon despite this step other times this precaution will save you from taking an unnecessary loss. Because of the uptick rule, it may take several attempts to execute a short trade. Dont be afraid to hit the short button on your trading platform software several times (assuming you are using a directaccess broker) so you can get in on an uptick. Check your trade confirmation window to make sure you are executing a single short trade, and not mistakenly ADBE had already reached the pre-determined entry price in pre-market trading. The stock kicked off the official trading session with a two-minute rally. Had we sold immediately on the open without adjusting the entry price to take this price action into account, we would have been stopped out with a loss. Adobe Systems Inc. (ADBE), two-minute 45.60 45.40 45.20 45.00 44.80 44.60 44.40 44.20 44.00 43.80 43.60 43.40 43.20 43.00 42.80 42.60 42.40 42.20 42.00 41.80 41.60 41.40 41.20 41.00 40.80 40.60 40.40 40.20 40.00 39.80 11:30 12:00 entering multiple trades. Because the stock already has traded at or close to this price in the pre-market, it also is important that the time and sales window confirms large block trades are going our way and that most trades are being executed at the bid price (indicating selling pressure). After the entry at 41.4, we trailed a stop a few spreads behind the open trade, without exceeding the 0.3-point stop weve set for this trade. Note that the stop is slightly tighter in this trade than in the first example. Because of the support-resistance level created by the pre-market gap to 41.6, this trade will be invalidated as soon as the market trades above this level, which will happen at 41.7 0.3 points away from the entry price. Most trades entered before 10 a. m. should not last any longer than five to eight minutes. Trades entered after 10 a. m. can last a little longer, but never more than 20 minutes. This trade was covered at 40.75 for a .65-point profit. Bottom line Successful trading is much more difficult than it first appears. It requires a long process of market watching and practicing chart pattern recognition. In time, you can learn to avoid low-potential situations and focus on entries based on specific chart pattern breakouts and breakdowns. Planning ahead to trade breakouts should be done daily using the previous days high and low to set trade alerts. Trading with the trend on breakouts using these criteria will help traders avoid overtrading and selectively trade the strongest and most powerful chart patterns. The only exceptions to trading breakouts of the previous days trading range are those rare occasions when a stock makes a rapid multi-point drop from the previous days high and bounces off the previous days low. But this is a trade for experienced traders only, and you should not expect to capture more than 50 percent of the retracements following the bounce. In fact, buying bottoms and shorting tops is largely a failing method, despite the amazing predisposition of most new traders to attempt these types of trades. Your trades should be at least 80 percent breakouts and no more than 20 percent bottom bounces, not the other way around. Its a good idea to tape that to your monitor, along with the words, Tight stops no exceptions Short signal is generated at 41.40. Previous days low: 42.20 Previous day (compressed) 9:30 5/3/01 10:00 Current day 10:30 11:00 5/2/01 Source: Data Broadcasting Corp. 34 activetradermag September 2001 ACTIVE TRADER TRADING Strategies Swing trading 10-day CHANNEL BREAKOUTS To trade breakouts successfully, you have to line up as many market factors as possible. Incorporating volume and momentum into your trading plan can put you on the inside track to breakout trades that wont break apart. BY KEN CALHOUN T 35 rying to outguess the market by picking bottoms and tops is usually unsuccessful, and more often than not results in a large numbers of whipsaw trades. By contrast, professional traders and institutions favor breakout trading. Combining 10-day support and resistance lines with confirming signals such as volume breakouts and reversals is a practical approach to identifying swing trade opportunities. These 10-day channels provide clear criteria for entering breakout trades once these price levels are triggered. Why swing trading Swing trading is a shorter-term trading style in which positions are held anywhere from one to 10 days. Swing trading has been increasingly popular ever since the Securities and Exchange Commission (SEC) raised the minimum margin requirement for pattern day traders (PDTs) to 25,000 on Sept. 28, 2001 (see New rules for the intraday trader, Active Trader, October 2001). Traders with less than the 25,000 can make no more than four intraday trades in a five-day period those who exceed this limit must meet the new day trading margin requirements or face potential position liquidation or account closure. Day trading, in which trades often are entered and exited in a matter of seconds, can be highly stressful and requires a significant initial investment not just in trading funds, but in computer hardware and software, and training in directaccess trading methods as well. Swing trading, by contrast, is generally less stressful and does not require as large an upfront investment in capital, software or equipment. Because it does not require a trader to watch the market all day, swing trading can be done on a part-time basis using online discount brokers. Professional day trading requires a full-time commitment and a fast direct-access broker. This makes swing trading a viable alternative for active traders who are unwilling to meet the new margin requirements and/or uncomfortable with the technology and capital demands of day trading. Swing trading is also an effective way to learn many of the classic technical indicators and limit risk with small-share or paper trades. The following strategy uses simple volume and sector-strength filters to determine when to trade breakouts of 10-day price channels. activetradermag March 2002 ACTIVE TRADER If a stock gaps The tools For this approach, use a 15-minute chart encompassing the most recent 10 days (i. e. today and the previous nine trading sessions). The following rules are given in terms of upside breakouts and long trades reverse the rules for short trades. However, this strategy is better for long swing entries. For a breakout swing trade, buy when a stock breaks out at least 50 cents over the whole number above the 10-day high, accompanied by volume that is higher than the previous days volume at the same time. For example, assume that during the previous nine trading sessions plus today, the highest price a stock traded at was 37.8, which was set on the previous trading day. The trigger for a 10-day breakout long trade would be 38.5, as long as the volume in the current session is higher than it was at the same time in the previous session. If the stock opened today at 37.6 and traded up to 38.5, this would trigger a long trade. The only exception is when the entry price would contain a 9 e. g. 19.5, 29.5, 39.5, and so on in such cases, wait until the stock clears the nearest multiple of 10, which would result in long trade triggers at 20.5, 30.5, 40.5, etc. The rationale is prices with a 9 tend to look expensive and often meet resistance, choppy price action, or both near multiples of 10. The rules The best types of stocks to trade with this approach are Nasdaq or NYSE stocks priced between 5 and 60, with average daily volume of at least 800,000 shares, and average intraday trading ranges of 1 to 4 points. Here are the rules: 1. Define the 10-day high and low for the stock using a 15-minute candlestick or bar chart. Be sure to include volume bars on the chart. 2. Enter 50 to 60 cents above the nearest whole number above the highest high of the past 10 days, including today. 3. Look for volume breakouts on the 10-day chart. Compare volume bars on the current trading day to previous trading days. The best entries are those for which volume is higher than in the previous session. 4. Confirm entries using market indicators such as the Arms Index (TRIN) as well as the time of day. The TRIN measures the net buying pressure vs. selling pressure in the market at a given point in the trading day. The formula is: / open more than 10 to 15 percent from its previous close, it will often reverse and fill the gap, in which case its necessary to take your profit before the market does. The TRIN, versions of which are available for both NYSE and Nasdaq stocks, can help determine whether a trade is advisable by highlighting whether momentum is bullish or bearish at a given time. A TRIN reading of 1 means buying/selling volume and the number of advancers/decliners are equally matched a TRIN reading above 1 is bearish a TRIN under 1 is bullish. See Indicator Insight, Active Trader, December 2000, for more information on this indicator. Its also helpful to enter at times of the day when the market is the strongest and most volatile. Its usually best to enter 10-day channel long trades in the early morning, from 9:45 until 11 a. m. EST, or during late-afternoon rallies between 2:30 and 3:30 p. m. for example. Certain cautionary indicators (red flags) can be used to eliminate poor trades. For long entries, avoid highs reached on lower-than-average volume or those reached by a stock in a weak sector that day. Compare sector indices such as the SOX, NBI, GHA and GSO to determine which are strongest, and give preference to entries in the strongest sectors 36 Strategy snapshot Strategy: 10-day channel breakout Markets: Nasdaq or NYSE stocks trading between 5-60, with average daily volume of at least 800,000 shares and average daily range of 1 to 4 points. Entry (for longs Go long 50 to 60 cents above the nearest whole number reverse for shorts): above the highest high of the past 10 days. Confirmation: The best entries are those in which volume is higher than in the previous session and/or when the stock is in a strongly trading sector. Avoid highs that are reached on lower-than-average volume or those reached by a stock in a weak sector on the entry day. Exit/risk control: The widest initial stop should be the closer of 1.5 points or the previous days low. The most conservative initial stop-loss is the previous days high. Once in a profitable trade, trail the stop .5 points below the current trading range. ACTIVE TRADER March 2002 activetradermag FIGURE 1 POISED TO BREAK OUT NVDA traded in a channel from 48 to 55 for 10 days, forming an ascending triangle toward the end of the period as it challenged the resistance level another time. Entry would occur at 55.50, 50 cents above the highest high of the past 10 days. Nvidia (NVDA), 15-minute Resistance 55.50 55.00 54.50 54.00 53.50 53.00 52.50 52.00 51.50 51.00 50.50 50.00 49.50 49.00 48.50 48.00 6 million 4 million 2 million 12:00 12:00 12:00 12:00 12:00 12:00 12:00 12:00 12:00 12:00 11/16/01 11/19/01 11/20/01 11/21/01 11/23/01 11/26/01 11/27/01 11/28/01 11/29/01 11/30/01 Ascending triangle Support Volume your risk tolerance. A good initial stop-loss is the previous days low or 1.5 points, whichever is smaller. For example, lets say the 10-day channel range for EBAY is bounded by a low (support level) of 61.8 and a high (resistance level) of 66.8. If the previous days range was 65.5 to 66.8, we would enter EBAY on a breakout above 67.5. The initial maximum stop-loss for this trade would be at 66, 1.5 points below entry (roughly 2 percent). If you trade on a shorter time frame, say three - or five-minute charts, you might consider setting a tighter stop at the previous days high. 6. Trail a stop to protect open profits at 2 percent (generally from .5 to 1.5 points) below the current level of the open trade, or use a time stop of no longer than 10 days (i. e. exit all remaining open positions after 10 days). Whenever one of the exit signals appears, the position should be closed with a profit. Re-enter on subsequent breakouts after retracements have occurred. Source: eSignal FIGURE 2 AFTER THE BREAKOUT After the stock fulfills the entry requirements and breaks out above resistance, a trailing stop is used to lock in profits. Intuit Inc. (INTU), 15-minute Entry Resistance 44.00 43.50 43.00 42.50 42.00 41.50 41.00 40.50 40.00 39.50 How to handle gaps Managing gap opens on swing trades is always a challenge. When a stock gaps open significantly above the previous days high (in your favor for a long swing trade), trail a stop no more than 50 cents below the current pre-market trading range to lock in your profit. However, if a stock gaps open more than 10 to 15 percent from its previous close, it will frequently reverse and fill the gap, in which case its necessary to take your profit before the market does. This is especially true if the stock gaps up above the previous days high. Conversely, when a stock gaps down significantly against you, its often best to wait until 15 to 20 minutes after the open to exit the position, because down gaps frequently attract buyers who can bring the price back up. Again, use a stop-loss of no more than 50 cents below the current pre-market trading range. It is sometimes helpful to wait until approximately 9:45 a. m. EST to see where the stock trades before exiting a position. It is frustrating to panic out of a gap-down swing trade only to see the stock turn around and fill the gap in the first few minutes of the trading day. Calmly give it a few minutes to establish a trend and see if it consolidates and Support Volume 39.00 600,000 400,000 200,000 12:00 12:00 12:00 12:00 12:00 12:00 12:00 12:00 12:00 12:00 11/16/01 11/19/01 11/20/01 11/21/01 11/23/01 11/26/01 11/27/01 11/28/01 11/29/01 11/30/01 Source: eSignal e. g. those up 1.5 to 3 percent or so on the day at the time of the trade entry. Also check to see if sectors are convergent (all green or red i. e. moving up or down) or divergent (mixed). Its best to enter swing trades on days where all sectors are convergent and the broad market has strength in a single direction. 37 Mixed, choppy days are poor days for swing trade entries. 5. Stop-loss values are determined by the previous days high and low. Either of these price points can provide you with an initial stop-loss value, depending on the intraday market trend and activetradermag March 2002 ACTIVE TRADER FIGURE 3 OVERFLOWING CUP Like channels and ascending triangles, cup patterns also provide well-defined resistance levels for breakout trades. The stock broke out above a second cup pattern on Dec. 7, and was stopped out with a 3.6 point profit on Dec. 11. Invision Technologies (INVN), 15-minute 40.00 38.00 36.00 34.00 Resistance Breakout entry Exit 32.00 30.00 28.00 26.00 24.00 Initial cup Second cup 22.00 20.00 18.00 16.00 Volume 1 million 500,000 12:00 12:00 12:00 11/29/01 11/30/01 12/3/01 12:00 12/4/01 12:00 12:00 12:00 12:00 12:00 12:00 12/5/01 12/6/01 12/7/01 12/10/01 12/11/01 12/12/01 Source: eSignal reverses this initial gap move. Trade examples Figure 1 shows Nvidia (NVDA) trading in a 10-day channel (between 48 to 55) from Nov. 16, 2001, to Nov. 30, 2001. Based on the guideline to enter 50 cents over the nearest higher whole number above the highest high of the 10 days, a long entry would be triggered at 55.5. (If, however, the stock gaps up to, say, 55.8 in pre-market trading, the entry would be reset over the next number up, at 56.5 or higher). The initial stop would be placed at 54, 1.5 points below the entry, because 1.5 points is a tighter stop than the previous days low (see trade rule No. 5). The alternate, more conservative stop-loss level is the previous days high in this case 54.60. Notice this stock forms an ascending triangle pattern following an initial downturn earlier in the 10-day channel, and is poised to break out to new highs if it clears the 55 resistance area. If the volume when the trade is entered (on Dec. 1, not shown) is higher than it was at the same time on the previous day, this would provide additional confirmation for a long trade. Figure 2 provides an example of how to manage a profitable long swing trade. On the afternoon of Nov. 29, 2001, the stock cleared the 10-day high and a long trade was entered at 42.50. The initial stop loss was set at 41 (42.5-1.5). On Nov. 30, the stock continued to rally throughout the session to a high of 44, at which point it consolidated. Using a trailing stop approach raises the stop to 43.5 (44-.5), which was not triggered. The time stop is 10 days from Nov. 29. In this case the stock is up more than a point on an overnight hold. We continue to trail a stop .5 behind the current trading range until the stop is taken out. it is sound risk management to extend yourself only on the strongest of patterns. This keeps the average entry price toward the low end of the total position. Cup-pattern breakouts, like ascending-triangle and consolidation breakouts, are much stronger than cases where a stock simply trades to a new high without penetrating any kind of resistance level. The test of sellers that occurs at a resistance level prior to a long cup-pattern breakout validates the entry and provides a support level after the trade. Cup patterns, which are extended, saucer-shaped retracements, appear fairly frequently. The key to trading them is to apply volume and other filters to avoid false breakouts that turn out to be double tops (i. e. when price falls back from the right side of the cup instead of breaking out above the resistance level of the cup). Figure 3 shows a cup pattern that started on Dec. 3, pulled back from the resistance level of 28.59 on Dec. 5 (forming a short-lived double top), formed another cup and finally broke out above the resistance on the afternoon of Dec. 7. A long trade was triggered at 30.5 (because the entry price would have been 29.5, and in such cases entry is made above the nearest multiple of 10). The stock gapped open higher on Dec. 10 and the trade was exited on Dec. 11 (when the trailing stop was hit) at 34.1 for a 3.6-point profit. Bottom line Swing trading provides traders with opportunities to manage multiple positions and entries at a more leisurely trading pace than is possible in the hectic world of day trading. However, every trader should research and experiment with different trading styles to help determine his or her preference and level of comfort. Using 10-day trading channels to identify entries on volume breakouts can help you better define support and resistance levels and provide techniques you can integrate with other technical indicators to develop a swing-trading plan. Using a comprehensive, measured and specific strategy to trade breakouts continues to produce entries that are more consistent than intra-range or bounce trade approaches. Using volume and price action filters will help you avoid false breakouts in choppy markets. 38 Dynamic position sizing using cup breakouts Dynamic position sizing is the process of adding to an initial position once a stock has broken out and is continuing to attract buyers. For example, a trader may buy 200 shares initially and add another 100 shares on a subsequent breakout as the stock continues to climb. The key to using dynamic position sizing is to add no more than half the number of the initial trade size on subsequent 10-day high cup-pattern breakouts. When adding to an initial position, ACTIVE TRADER March 2002 activetradermag EQUITY CURVE 3,000,000 2,500,000 System logic: The volatility breakout system is a classic trading strategy based on identifying situations when a 1,500,000 market is about to burst out of a congestion area and potentially establish a new long-term trend. It also can signal a 1,000,000 trade if the market is already trending in one direction but quickly reverses to establish a new trend in the opposite 500,000 direction. This system uses Bollinger Bands (complemented by 0 moving averages), which are lines typically plotted two 12/7/92 12/7/93 12/7/94 12/7/95 12/7/96 12/7/97 12/7/98 12/7/99 12/7/00 12/7/01 standard deviations above and below a moving average. Bollinger Bands expand during high-volatility periods and contract during low-volatility periods. Test data: Daily prices for 14 Dow Jones Industrial Average stocks When volatility is high, the system is designed to stay out of the (AXP, C, CAT, DIS, GM, HWP, IBM, INTC, JPM, KO, MO, MRK, market to avoid taking any unnecessary risk, but if an entry is trig - MSFT, T), with 10 deducted per trade for slippage and commisgered anyway, the system will work to keep you in the trade to sions. avoid being stopped out prematurely with a loss. A long entry is triggered when price moves above its 60-day moving average and Starting equity: 1 million (nominal). breaks the upper Bollinger Band. To exit, price must move below its 30-day moving average and break a lower Bollinger Band that is set Buy-and-hold stats: one standard deviation away (instead of the usual two). Because the Total Maximum Longest next entry cannot occur until price moves back above the lower Index return drawdown flat period band, above its moving average and above the upper band, there DJIA 175 31.5 (current) 29 months (current) will be times when the system is out of the market completely. SP 500 120 40 (current) 26 months (current) Nasdaq 182 80.5 (current) 26 months (current) Markets: This system will be tested on stocks and also on futures (p. 70). Test results: The system was originally tested on all 30 stocks in the DJIA the 14 in this test were selected because they were the Rules: ones that showed a profit. Singling out these stocks, though, 1. Go long tomorrow if price moves above its average price for the last 60 days and breaks the SAMPLE SIGNALS upper Bollinger Band. 2. Exit with a profit or loss if 58.00 Philip Morris (MO), daily price moves below its 30-day moving average and penetrates LX Long exit 56.00 LX a lower Bollinger Band set one standard deviation away. LX Buy Sell 52.00 54.00 Reverse the rules for short trades. Money management: 1. Risk 4 percent of available equity per stock traded. 2. The number of shares to trade (ST) is calculated using the following formula: ST AC PR / R where AC Available capital PR Percent risked R Distance between entry price and exit price (stop-loss). Test period: November 1992 to June 2002. 39 February Source: Omega Research ProSuite March April May June Account balance () Volatility breakout system 2,000,000 50.00 Buy 48.00 46.00 44.00 42.00 July activetradermag October 2002 ACTIVE TRADER DRAWDOWN CURVE 12/7/92 0 -5 -10 -15 -20 -25 -30 12/7/93 12/7/94 12/7/95 12/7/96 12/7/97 12/7/98 12/7/99 12/7/00 12/7/01 failed to create results as good as the ones produced by the test on currency futures (see Futures System Lab, p. 44). One thing to keep in mind is that portfolio composition is more complex than simply eliminating instruments that dont perform well. In a dynamic portfolio such as this one, where a group of stocks interacts within a single trading account, the ability of an individual stock to turn a profit or loss also depends on the behavior of all the other stocks. As proof of this, three stocks that were profitable when all 30 stocks were tested showed a loss when the field was pared to 14. If we were to test only the remaining 11 that showed a profit, its likely a few more would turn into losers. This correlation also makes it extremely difficult to trade a longterm system on the stock market. The individual stocks either trend well, almost all at once, or they dont, which results in a large amount of whipsaw losing trades and rather severe drawdowns. The way to overcome this is to diversify by trading as many stocks from different sectors and groups as possible. In this system, though, the large distance between the entry and exit prices requires trading rather small positions. Doing otherwise would run the risk of using all the capital on only a few positions. This in turn will result in relatively small dollar gains despite large price swings. Even though the system trades only 14 stocks, close to 75 percent of available capital is tied up. Also, the current drawdown has gone on for 42 months. This is likely a reflection of the disappearance of the stock markets pre-2000 trending characteristics, and it is not very likely that a system like this will start producing a profit anytime soon. This system is also a bit passive in its trade frequency. To make it more aggressive, the lookback period can be shortened and/or the standard deviation boundaries tightened. ROLLING TIME WINDOW RETURN ANALYSIS Cumulative Most recent: Average: Best: Worst: St. dev. Annualized Most recent: Average: Best: Worst: St. dev: 12 months 16.70 10.76 58.88 -15.08 17.29 12 months 16.70 10.76 58.88 -15.08 17.29 24 months 36 months 48 months 60 months -0.91 -5.86 15.22 22.04 24.50 42.81 64.40 83.77 87.95 102.78 162.29 155.19 -19.33 -16.50 4.57 16.53 22.72 30.43 36.41 42.26 24 months -0.45 11.58 37.10 -10.19 10.78 36 48 months months -1.99 12.61 26.57 -5.83 9.26 3.61 13.23 27.26 1.12 8.07 60 months 4.06 12.94 20.61 3.11 7.30 STRATEGY SUMMARY Profitability End. equity (): 2,491,172 Total return (): 149 Avg. annual ret. (): 10.00 Profit factor: 1.34 Avg. tied cap (): 73 Win. months (): 53 Drawdown Max. DD (): Longest flat (m): 25.8 41.5 Trade statistics No. trades: 456 Avg. trade (): 3,270 Avg. DIT: 35.0 Avg. win/loss (): 31,090 (13,546) Lrg. win/loss (): 391,627(109,437) Win. trades (): 38.8 TIM (): Tr./Mark./Year: Tr./Month: 100 57.7 4.3 4.0 LEGEND: Cumulative returns Most recent: most recent return from start to end of the respective periods Average: the average of all cumulative returns from start to end of the respective periods Best: the best of all cumulative returns from start to end of the respective periods Worst: the worst of all cumulative returns from start to end of the respective periods St. dev: the standard deviation of all cumulative returns from start to end of the respective periods Annualized returns The ending equity as a result of the cumulative returns, raised by 1/n, where n is the respective period in number of years LEGEND: End. equity () equity at the end of test period Total return () total percentage return over test period Avg. annual ret. () average continuously compounded annual return Profit factor gross profit/gross loss Avg. tied cap () average percent of total available capital tied up in open positions Win. months () percentage profitable months over test period Max. DD () maximum drop in equity Longest flat longest period, in months, spent between two equity highs No. trades number of trades Avg. trade () amount won or lost by the average trade Avg. DIT average days in trade Avg. win/loss () average wining and losing trade, respectively Lrg. win/loss () largest wining and losing trade, respectively Win. trades () percent winning trades TIM () amount of time there is at least one open position for entire portfolio, and each market, respectively Tr./Mark./Year trades per market per year Tr./Month trades per month for all markets Send Active Trader your systems If you have a trading system or idea youd like tested, send it to us at the Trading System Lab. Well test it on a portfolio of stocks or futures (for now, maximum 60 markets, using the last 2,500 trading days), using true portfolio analysis/optimization. Most system-testing software only allows you to test one market at a time. Our system-testing technique lets all markets share the same account and is based on the interaction within the portfolio as a whole. Start by e-mailing system logic (in TradeStations EasyLanguage or in an Excel spreadsheet) and a short description to editorialactivetradermag, and well get back to you. Note: Each system must have a clearly defined stop-loss level and a suggested optimal amount to risk per trade. Disclaimer: The Trading System Lab is intended for educational purposes only to provide a perspective on different market concepts. It is not meant to recommend or promote any trading system or approach. Traders are advised to do their own research and testing to determine the validity of a trading idea. Past performance does not guarantee future results historical testing may not reflect a systems behavior in real-time trading. ACTIVE TRADER October 2002 activetradermag 40 FUTURES OPTIONS System Lab EQUITY CURVE 5,000,000 4,500,000 4,000,000 Futures volatility breakout system System logic: The system uses Bollinger Bands and a moving average to trade volatility breakouts. The logic and tools for this system are described in the Trading System Lab on p. 56. Markets: Most trending futures markets, such as currencies, energies and interest rates. This system was also tested on stocks (see Trading System Lab). Rules: 1. Go long tomorrow if price moves above the 60-day moving average and breaks the upper Bollinger Band. 2. Exit with a profit or loss if price moves below its 30-day moving average and penetrates a lower Bollinger Band set one standard deviation away (instead of the usual two). Reverse the rules for short trades. Money management 1. Risk the following percentages of available equity per market: 2 percent for Australian dollar, British pound, Canadian dollar, Dollar index and Swiss franc, and 4 percent for Japanese yen, D-mark and Euro. (The yen and the Euro are traded with twice the risk because they are the most liquid currencies.) 2. The number of contracts to trade (CT) is calculated with the following formula: CT (AC PR) / (R PV) where AC Available capital PR Percent risked R Distance between entry price and exit price (stop-loss). PV Dollar value of a one-point move. Test period: November 1992 to June 2002 Test data: Daily futures prices for eight currency futures: Japanese yen, Australian dollar, Canada dollar, British pound, Dollar index, Swiss franc, and D-mark (until Dec. 31, 1999)/Euro (after Dec. 31, 1999). Starting equity: 1 million (nominal). Test results: If there ever was a system built for the currency markets, this is it. The reason is the smooth, long-term trends currencies sometimes exhibit, probably because they are mostly influenced by the global, long-term economical and political climate. In contrast, the stock market is very sensitive to all other markets while the stock market cares a great deal about the currency market, the currency market doesnt care all that much about the stock market. Account balance () 3,500,000 3,000,000 2,500,000 2,000,000 1,500,000 1,000,000 500,000 0 11/25/92 11/25/93 11/25/94 11/25/95 11/25/96 11/25/97 11/25/98 11/25/99 11/25/00 11/25/01 The risk for this system was 2 or 4 percent per trade, which resulted in both a drawdown and flat time that would be deemed unacceptable by most professional money managers. These numbers should stay under 30 percent and 18 months, respectively. The major disadvantage of a trend-following system is most markets that work well with this type of a system are usually correlated. In other words, they will either all work well or perform poorly simultaneously. This is reflected in the erratic look of the systems equity curve. Because of this, the drawdowns can be both deep and long, and it usually requires a couple of very good trades to get the system profitable again. Research has shown that a trend-following system will work best when traded on 15 to 20 select markets from various sectors of the economy. The system ties up an average of 11 percent of capital. This means there is plenty of room to add markets, and even trade the system more aggressively, without extending ourselves too much. This is because futures have much smaller margin requirements than stocks. Theoretically, as many as 40 to 50 different futures contracts could be traded before reaching the same level of margin fewer than 20 stocks would require. Finally, the most recent drawdown of approximately 30 percent is the only one over the last 10 years of such magnitude. Most of the previous drawdowns bottomed between 10 to 20 percent. Therefore, the latest drawdown is quite possibly an anomaly. That said, however, there are no guarantees in the market, and your worst drawdown is always still to come. The system also has a relatively low trade frequency. To make it more aggressive, the lookback period can be shortened and/or the standard deviation boundaries tightened. ROLLING TIME WINDOW RETURN ANALYSIS Cumulative 12 24 36 48 60 months months months months months Most recent: 34.31 58.39 65.73 114.82 138.14 Average: 15.07 34.15 55.68 80.53 109.16 Best: 74.75 76.50 108.66 137.94 202.94 Worst: -28.10 -1.33 4.12 28.76 38.77 St. dev. 18.17 16.26 23.27 22.74 37.29 Annualized 12 24 36 48 60 months months months months months Most recent: 34.31 25.85 18.34 21.06 18.95 Average: 15.07 15.82 15.90 15.91 15.90 Best: 74.75 32.86 27.78 24.20 24.82 Worst: -28.10 -0.67 1.36 6.52 6.77 St. dev: 18.17 7.82 7.22 5.26 6.54 LEGEND: Cumulative returns Most recent: most recent return from start to end of the respective periods Average: the average of all cumulative returns from start to end of the respective periods Best: the best of all cumulative returns from start to end of the respective periods Worst: the worst of all cumulative returns from start to end of the respective periods St. dev: the standard deviation of all cumulative returns from start to end of the respective periods Annualized returns The ending equity as a result of the cumulative returns, raised by 1/n, where n is the respective period in number of years STRATEGY SUMMARY Profitability End. equity (): 4,550,290 Total return (): 355 Avg. annual ret. (): 17.13 Profit factor: 1.32 Avg. tied cap (): 11 Win. months (): 51 Drawdown Max. DD (): 31.6 Longest flat (m): 19.7 Trade statistics No. trades: 303 Avg. trade (): 11,717 Avg. DIT: 35.8 Avg. win/loss (): 53,057 (29,536) Lrg. win/loss (): 345,600 (131,350) Win. trades (): 42.1 TIM (): 97 54.1 Tr./Mark./Year: 4.0 Tr./Month: 2.6 LEGEND: End. equity () equity at the end of test period Total return () total percentage return over test period Avg. annual ret. () average continuously compounded annual return Profit factor gross profit/gross loss Avg. tied cap () average percent of total available capital tied up in open positions Win. months () percentage profitable months over test period Max. DD () maximum drop in equity Longest flat longest period, in months, spent between two equity highs No. trades number of trades Avg. trade () amount won or lost by the average trade Avg. DIT average days in trade Avg. win/loss () average wining and losing trade, respectively Lrg. win/loss () largest wining and losing trade, respectively Win. trades () percent winning trades TIM () amount of time there is at least one open position for entire portfolio, and each market, respectively Tr./Mark./Year trades per market per year Tr./Month trades per month for all markets 41 activetradermag October 2002 ACTIVE TRADER TRADING Strategies Better breakout trading: THE NOISE CHANNEL SYSTEM All breakout traders have to deal with the reality of false moves BY DENNIS MEYERS, PH. D. and whipsaws. The noise channel breakout system shows how a filter can improve the performance of intraday breakout trading. In the tests that illustrate this strategy, well use five-minute bars of IBM from Feb. 21 to April 6. (For an important point on testing stock trading strategies, see A note on price data and dividends, p. 75). Intraday data has a high noise level, meaning it contains a great deal of random price movement that looks significant but turns out to be meaningless. Without some kind of filter, the losses generated by the random price movement (that is, whipsaws) can completely overwhelm a trading system. To help eliminate such random movement, we will add a noise filter, designated by the symbol f, to the basic channel breakout system. There are three system parameters to find: nhi, which is the number of bars in the lookback period used to determine the highest high price (hhp). nlo, which is the number of bars in the lookback period used to determine the lowest low price (llp). f, which is the amount price must exceed the hhp or llp to trigger a buy or sell. The rules for the resulting noise channel breakout system (NCBS) are: Buy rule: If price crosses above the highest high of the last n rice trends begin with a breakout of a previous high or previous low. Unfortunately, many breakouts are random mere market noise. False moves and reversals can repeatedly whipsaw traders who act immediately on typical breakout signals. As a result, traders sometimes attempt to use filters to improve the odds of catching a successful breakout trend. One example of a simple filter is to wait for consecutive closes above or below a breakout level. Another example is waiting for price to penetrate a breakout level by x percent or points before acting on the signal. The following discussion will analyze a variation on the simple channel breakout system that uses the latter type of filter to minimize whipsaws on an intraday basis. The strategy will be tested on International Business Machines (IBM). The discussion is broken into two parts, covering 1) the system rules and data selection and 2) testing procedures. This will give you the necessary tools for performing similar research and tests on other markets. P The noise channel breakout system The basic system we will use here is a fairly simple and effective breakout system that has been in the public domain for many years: the channel breakout system, which goes long on a move above the highest high of the last n bars and goes short on a move below the lowest low of the last n bars. bars (nhi) by an amount greater than or equal to f, buy at market. For example, if n 20 and f 2 (points), you would go long when price moved 2 points above the highest high of the last 20 bars. In addition, when short, and when calculating the highest high price (hhp), it cannot be higher than the previously calculated hhp as previous highs are dropped out of the lookback window. Otherwise, a situation can occur where there is a higher hhp without the price filter f being hit. Therefore, when short the stock, the hhp can only stay the same or go lower. It cannot go higher. Sell rule: If price crosses below the lowest low price of last n bars (nlo) minus an amount greater than or equal to f, sell at market. In addition, when long and when calculating the lowest low price (llp), it cannot be lower than the previous calculated llp as previous lows are dropped out of the lookback window. Again, to avoid the situation where a lower llp occurs without the price filter f being hit, when long the stock, the llp can only stay the same or go higher. It cannot go lower. Exit rule: Close the position five minutes before the NYSE close (no trades are carried overnight). stable i. e. the profits, wins and drawdowns should not change much as the parameters move by a small amount away from their optimum values. In other words, the system TABLE 1 OPTIMUM PARAMETER VALUES FOR TEST DATA Start date 2/21/01 2/28/01 End date 3/23/01 3/30/01 nhi 8 8 nlo 4 4 f 1 1 TABLE 2A TEST PERIOD 1 Performance summary for noise channel breakout system: IBM, five-minute bars from Feb. 21 to March 23. Statistics based upon trading 1,000 shares of IBM. Performance summary: Total net profit (): Gross profit (): Total no. of trades: Number winning trades: Largest winning trade (): Average winning trade (): Ratio avg. win/avg. loss: Max. consec. winners: Avg. Nein. bars in winners: Max intraday drawdown (): Profit factor: 13,890 39,260 48 26 5,940 1,510 1.309 4 39 -8,470 1.547 Max. Nein. contracts held: 1 All trades Open position P/L (): Gross loss (): Percent profitable (): Number losing trades: Largest losing trade (): Average losing trade (): Avg. trade(win loss) (): Max. consec. losers: Avg. Nein. bars in losers: 0 -25,370 54 22 -2,060 -1,153.18 289.38 3 21 Testing the strategy The walk-forward testing approach will be used to test this strategy because of the volatile nature of intraday stock prices. Intraday price dynamics are constantly changing because of economic surprises, events and trader sentiment. Also, the time of year such as the season, holidays, vacation time, etc. affects the character of intraday markets. As a result, tests performed on intraday data three months ago may no longer be representative of todays intraday price action. For more information on walkforward testing and how it was used for this strategy, see Proper system testing,. The best parameters will be defined as those values that generate the best net profits combined with the minimum drawdown and minimum largest losing trades. In addition, the results should be 43 TABLE 2B TEST PERIOD 2 Performance summary for noise channel breakout system: IBM, five-minute bars from Feb. 28 to March 30. Statistics based upon trading 1,000 shares of IBM. Performance summary: Total net profit (): Gross profit (): Total no. of trades: Number winning trades: Largest winning trade (): Average winning trade (): Ratio avg. win/avg. loss: Max. consec. winners: Avg. Nein. bars in winners: Max. intraday drawdown (): Profit factor: 10,490 35,500 47 22 5,940 1,613.64 1.613 3 39 -9,660 1.419 Max. Nein. contracts held: 1 All trades Open position P/L (): Gross loss (): Percent profitable (): Number losing trades: Largest losing trade (): Average losing trade (): Avg. trade(win loss) (): Max. consec. losers: Avg. Nein. bars in losers: 0 -25,010 47 25 -1,840 -1,000.40 223.191 3 26 Source: TradeStation by TradeStation Group Inc. activetradermag September 2001 ACTIVE TRADER performance using an nhi of 10 bars should be similar to that using nine bars or 11 bars. Also, in choosing the best parameters, we considered only those results with four or less maximum consecutive losses. from March 26 to April 6. The trades in this time period are the out-of-sample trades TABLE 3 OUT-OF-SAMPLE RESULTS generated from the optimized parameters from the two test sections of Feb 21 to Test results Table 1 shows the optimum parameter values for the test window described in Proper system testing. The nhi was eight bars, the nlo was four bars and f was 1 point. Tables 2a and 2b show test results using these parameters. Table 3 summarizes the combined performance of the two out-of-sample data segments from March 26 to April 6. This performance represents what would have happened in real time if you used the system parameters found in the test section (not including slippage and commissions). By comparison, the same nhi and nlo values tested without any filter resulted in a loss of 1,150. Table 4 is a trade-by-trade summary TABLE 4 TRADE-BY-TRADE SUMMARY Combined walk-forward out-of-sample performance summary for the noise channel breakout system: IBM five-minute bars from March 26 to April 6. Statistics based upon trading 1,000 shares of IBM. Performance summary: Total net profit (): Gross profit (): Total no. of trades: Number winning trades: Largest winning trade (): Average winning trade (): Ratio avg. win/avg. loss: Max. consec. winners: Avg. Nein. bars in winners: Max. intraday drawdown (): Profit factor: 8,390 14,460 16 8 4,000 1,807.50 2.382 5 54 -4,480 2.382 Max. Nein. contracts held: 1 All trades Open position P/L (): Gross loss (): Percent profitable (): Number losing trades: Largest losing trade (): Average losing trade (): Avg. trade(win loss) (): Max. consec. losers: Avg. Nein. bars in losers: 0 -6,070 50 8 -1,350 -758.75 524.38 3 37 Source: TradeStation by TradeStation Group Inc. This trade summary for the out-of-sample test (five-minute bars, March 26 to April 6) of the noise channel breakout system shows the strategy actually worked better on the short side than the long side. Entry Date Entry time Buy or sell Sell Buy Sell Buy Sell Buy Sell Buy Sell Sell Buy Sell Buy Sell Buy Sell Entry price 93.75 95.59 97.92 96.05 94.90 96.70 96.20 97.75 96.40 93.00 92.00 92.00 95.68 97.30 98.24 97.30 Exit date 3/26/01 3/27/01 3/28/01 3/29/01 3/29/01 3/30/01 3/30/01 4/2/01 4/2/01 4/3/01 4/4/01 4/4/01 4/5/01 4/6/01 4/6/01 4/6/01 Exit time 15:55 15:55 15:55 15:05 15:55 13:05 15:55 10:55 15:55 15:55 13:50 15:55 15:55 11:55 12:35 15:55 Exit price 94.52 99.59 94.50 94.90 94.88 96.20 96.25 96.40 94.50 90.50 92.00 91.85 98.15 98.24 97.30 97.67 bars in trade 67 68 75 60 10 41 34 15 60 71 49 25 75 27 8 40 PL () PL () Max. profit 0 4,300 3,420 950 390 800 220 350 2,600 2,740 1,900 380 3,040 550 1,660 300 Time Max. drawdown () (1,620) 0 (380) (1,160) (500) (1,190) (840) (1,350) (1,300) 0 (1,890) (500) (10) (940) (940) (1,960) Time 3/26/01 10:20 3/27/01 10:15 3/28/01 9:40 3/29/01 10:05 3/29/01 15:05 3/30/01 4/2/01 4/2/01 4/3/01 4/4/01 4/4/01 4/5/01 4/6/01 4/6/01 4/6/01 9:40 9:40 10:55 10:00 9:45 13:50 9:40 9:40 11:55 12:35 3/30/01 13:05 (770) 4,000 3,420 (1,150) 20 (500) (50) (1,350) 1,900 2,500 0 150 2,470 (940) (940) (370) -0.82 4.18 3.49 -1.20 0.02 -0.52 -0.05 -1.38 1.97 2.69 0.00 0.16 2.58 -0.97 -0.96 -0.38 10:20 15:50 12:00 10:30 15:15 11:55 13:15 10:05 15:40 15:40 11:20 14:00 15:25 11:15 12:05 12:35 10:35 10:15 9:40 11:20 15:45 10:00 14:35 10:55 11:40 10:00 10:30 14:20 9:40 11:55 12:35 13:55 ACTIVE TRADER September 2001 activetradermag 44 A note on price data and dividends A n overlooked aspect of testing a stock trading strategy is the effect of dividends. For example, IBM pays dividends on a quarterly basis, usually on the dividend payable dates of March 10, June 10, Sept. 10 and Dec. 10. On the Ex-dividend dates (approximately one month before the payable date), the price of the stock is adjusted down by the value of the dividend. Thus, over the course of a year, IBM has a small downward price bias equal to the amount of the yearly dividend. If you were an owner of IBM, you would receive those dividends in cash, making up for the small downward bias. However, when developing and testing a system using historical stock data, prices are not adjusted for dividend payments. This creates a small distortion in parameter selection and forward-adjusted results. Because no dividends were paid in the data sample used for the test in this article, no adjustment needs to be made. However, if the intraday time period fell on an ex-dividend date, an adjustment would have to be made to avoid distortion. March 23 and Feb. 28 to March 30. Figure 1 is a five-minute chart with the noise channel superimposed, as well as some of the buy and sell signals from the Table 4 trade-by-trade summary. Breaking down the numbers With respect to average winning and losing trades, drawdowns and profit factor, the out-of-sample performance (Table 3) was better than the test sample performance (Tables 2a and 2b) The better performance of the out-of-sample section could have been coincidental, but it does indicate that four weeks of test data was enough to capture the intraday price dynamics of this stock. Proper system testing hen testing any trading strategy, the important point is how well it will perform on price data it has not been optimized on that is, out-of-sample data. In short, without out-of-sample testing, its nothing more than a hope and a prayer to believe that system performance in the future will be anywhere near the optimized performance. For example, its possible to take a trading strategy with four independent variables, or parameters, and with hindsight, find the values for each of them that give the best (optimized) results on a specific historical period say, the last three years (using daily price data). However, these optimized parameter values have, in essence, been cherry-picked for this particular data period (a process known as curve-fitting), and are unlikely to perform as well on other historical test periods, or in actual trading in the future. A walk-forward testing procedure was applied to the noise channel breakout system as follows: Five-minute bars from a period of four weeks from the start of the test period Feb. 21 to March 23 were chosen and system parameter values were found through optimization on this intraday data segment. In other words, the best-performing system parameters (e. g. number of days in lookback period, noise filter value) were determined by testing a range of values for each. At this stage of system development, the only thing indicated by the optimum values in the test portion is that the data has been curve-fitted as best it can with this system. Without further testing on out-of-sample data, there is no way to tell if the system will work in the future. W These parameter values were then applied to an out-ofsample data period following the test segment (March 26 to March 30). This walk-forward process was repeated by moving the test data window forward one week, to Feb. 28 to March 30, and again finding the parameters values through optimization on this new data. These optimized parameter values are then applied to the next out-of-sample five-minute intraday data window (April 2 to April 6). An important (but unspoken) point in walk-forward testing is that if you cannot get good results in the out-of-sample data segments, real-time system performance will be random. Almost any period of historical prices can be curve fitted easily to give the false illusion of future profitability. However, these performance measures in no way reflect how a system will perform on price data it has not been optimized on. Only out-of-sample testing that is, testing on price data the system parameters were not originally derived from can determine if a system is robust and has a chance of performing well in real trading. Despite these facts, many market pundits still make the unproven claim that statistics generated solely from optimized buy and sell trades in the test section (the initial period of price data)have value in predicting whether or not the system will perform well in the future. Nothing could be further from the truth. The only thing the statistics from the test section tell you is how well you have curve-fitted the data in the test section. As a matter of fact, using optimization, its almost impossible not to get an excellent fit with great statistical results. 45 activetradermag September 2001 ACTIVE TRADER The out-of-sample trade summary (Table 4) shows the system did better on short trades than it did on long trades. On one hand, this could indicate a negative bias for the system. On the other hand, given the current bear market environment, the ability to cash in on the short side has value. There were no big winners or big losers, indicating steady returns. Average wins were 2.4 times average losses in the out-of-sample section. Figure 1 shows how the system was able to efficiently capture intraday trends in IBM. Also, the system constraint of not carrying positions overnight eliminated many negative opening surprises. Overall, traded on IBM, the NCBS did a good job of minimizing the whipsaw losses prevalent in breakout trading systems and maximizing the profits from major intraday trend moves. Building on the results To use this system in real-time trading, at least 10 additional test and out-of-sample windows should be examined to ensure the performance summarized here was not the result of chance. To determine if this approach can be used on other stocks or markets it would be necessary to follow the same proce - dures and determine the appropriate parameters for each. Every market has subtle differences because the participants vary from market to market. Also, market activity can change over time. Consequently, you should continue to perform walk-forward testing to determine if there is a shift in the systems effectiveness and whether better parameters have emerged. Any trading method should be tested before you risk capital on the technique. Granted, there is a considerable amount of work involved, but without taking the time to adequately research a technique, the chances of success are low. FIGURE 1 NOISE CHANNEL AND TRADE SIGNALS A few of the buy and sell signals generated by the noise channel breakout system are shown. The system successfully captured intraday trends. International Business Machine (IBM), five-minute 101 100 -1 99 98 97 96 95 1 94 93 3/26 11:20 12:15 1:10 2:05 3:00 3/27 11:15 12:10 1:05 2:00 2:55 3/28 11:10 12:05 1:00 1:55 2:50 Source: TradeStation by TradeStation Group ACTIVE TRADER September 2001 activetradermag 46 TRADING Strategies The long and short of it: The Noise Channel Breakout SYSTEM 2 BY DENNIS MEYERS, PH. D. A lthough price breakouts are the basis for many trading approaches, breakout systems are plagued by false signals when price initially breaks out, triggering a buy or sell, but quickly retraces, resulting in a losing trade. To combat this problem, traders often apply filters to breakout systems, delaying trade entry until the initial breakout has been confirmed by a price move of a certain size or duration in the direction of the breakout. Better breakout trading: The noise channel breakout system (Active Trader, September 2001, p. 70) showed how a simple channel breakout system, with an additional noise filter to minimize whipsaws, could be developed to trade IBM on an intraday basis using five-minute bars. The noise filter delays taking a breakout signal until the market provides some confirmation the breakout is sustainable, thus avoiding false breakouts. One aspect the original noise channel breakout system (NCBS) did not address is whether to treat the long and short sides of the market the same that is, whether the filter should be different for long and short trades, since uptrends and downtrends have different characteristics. Here we will use the NCBS, again applied to IBM fiveminute price bars, to see if some improvement can be made by using different filters for long and short trades, respectively. To compare the new version of the system to the previous one, the following tests will use the same five-minute bar prices of IBM from Feb. 21, 2001, to April 6, 2001. First, well review the basics of breakout systems in general and the NCBS in particular. NCBS refresher The basic channel breakout system goes long on a move above the highest high of the last n bars and goes short on a move below the lowest low of the last n bars. For example, a 40-day channel breakout goes long when price moves above the highest high of the last 40 days and goes short when price falls below the lowest low of the last 40 days. Breakout systems can be used on intraday price data, as well as daily or weekly data. The NCBS is an intraday breakout system based on five-minute bars. Because intraday price action can be very volatile, without some kind of filter the losses generated by the random price movement (that is, whipsaws) can completely overwhelm a trading system. To help eliminate activetradermag October 2001 ACTIVE TRADER 47 Traders often use additional rules or filters to prevent being whipsawed by breakout trading strategies. Find out if using different filters for long and short trades improves the performance of an intraday breakout strategy. WALK-FORWARD: Proper system testing hen testing any trading strategy, the important point is how well it will perform on data on which it has not been optimized that is, out-of-sample data. If a certain system is first tested on a sample piece of historical price data (say, daily bars from 1993 up to 1998), the systems performance results have no implication outside this sample data set all you know is how well your system parameters performed during this particular period. To get an idea of how the system might actually perform, the system parameters used for the sample data should be applied to different out-of-sample price data (say, daily bars from 1998 to the present). This walk-forward process simulates the application of a system to future data, as would occur in actual trading. In short, without out-of-sample testing, its nothing more than hope to believe that system performance in the future will be anywhere near the optimized performance. For example, its possible to take a trading strategy with four independent variables, or parameters, and with hindsight, find the values for each of them that give the best (optimized) results on a specific historical period say, the last three years (using daily price data). However, these optimized parameter values have been, in essence, cherry-picked for this particular data period (a process known as curve-fitting), and are unlikely to perform as well on other historical test periods, or in actual trading in the future. An important (but unspoken) point in walk-forward testing is that if you cannot get good results in the out-ofsample data segments, real-time system performance will be random. W such random movement, the NCBS adds a noise filter, designated by the symbol f, to the basic channel breakout system. The three system parameters for the NCBS are: nhi, which is the number of bars in the lookback period used to determine the highest high price (hhp). nlo, which is the number of bars in the lookback period used to determine the lowest low price (llp). f, which is the amount price must exceed the hhp or llp to trigger a buy or sell. The Noise Channel Breakout System 2 The Noise Channel Breakout System 2 (NCBS-2) uses different filter values (f, from the original system) for the long and short sides of the market. As a result, there are four system parameters for the NCBS-2: nhi, which is the number of bars in the lookback period used to determine the highest high price (hhp). nlo, which is the number of bars in the lookback period used to determine the lowest low price (llp). ACTIVE TRADER October 2001 activetradermag 48 Think of the symbols xoU and xoD as the crossover Up and crossover Down values. The logic behind modifying the filter values is because market behavior associated with buys and sells is quite different, TABLE 1 OPTIMUM PARAMETER VALUES FOR TEST DATA the noise channels associated with buys and sells should be independent of each other. The NCBS-2 rules are: Buy rule: If price crosses above the highest high price of the Start date End date nhi xoU nlo xoD last nhi bars by an amount greater then or equal to xoU, then buy at the market. In addition, when short, and when calculat2/21/01 3/23/01 8 1 4 1 ing the highest high price (hhp), the hhp can only stay the same or go lower than its most recent value, it cannot go higher. 2/28/01 3/30/01 18 1.25 12 0.25 Sell rule: If price crosses below the lowest low price of last nlo days by an amount of greater than or equal to xoD, then sell at the market. In FIGURE 1A TEST PERIOD 1 addition, when long and when calculating the lowest low price (llp), the llp can Performance summary for NCBS-2, IBM five-min. bars, Feb. 21 to March 23. only stay the same or go higher than its Statistics based upon buying and selling 1,000 shares of IBM. most recent value, it cannot go lower. Exit rule: Close any position five minPerformance summary: All trades utes before the New York Stock Exchange Total net profit (): 13,890 Open position P/L (): 0 close (no trades are carried overnight). Gross profit (): 39,260 48 26 5,940 1,510 1.309 4 39 -8,470 1.547 Max. Nein. contracts held: 1 Gross loss (): -25,370 54 22 -2,060 -1,153.18 289.38 3 21 Total no. of trades: Number winning trades: Largest winning trade (): Average winning trade (): Ratio avg. win/avg. loss: Max. consec. winners: Avg. Nein. bars in winners: Max intraday drawdown (): Profit factor: Percent profitable (): Number losing trades: Largest losing trade (): Average losing trade (): Avg. trade(win loss) (): Max. consec. losers: Avg. Nein. bars in losers: xoU, which is the amount price must exceed the hhp to trigger a buy signal. xoD, which is the amount price must fall below the llp to trigger a sell signal. Testing the strategy As in last months article, walk-forward optimization will be used here. The same data will also be used so we can judge whether this new modification can improve performance. The walk-forward testing procedure was applied as follows: A four-week period from the start of the IBM five-minute bar data from Feb. 21 through March 23 was chosen and system parameter values were found through optimization on this intraday data segment. (Optimization refers to the search for the parameter values that give the best results.) It should be noted that in this stage of system development, the only thing indicated by the optimum values that are found in the test portion is that the data has been curve fitted as best it can with this system. Without further testing on out-of-sample data, there is no way to tell if the system will work in the future. The parameter values found were then applied to an out-of-sample period, in this case March 26 to March 30. This process was repeated by moving the test data window forward one week to Feb. 28 through March 30 and again finding the parameters values through optimization on this new data test window. The parameter values found were then applied to the next out-of-sample data set, which in this case was April 2 to April 6. See Walkforward: Proper system testing for addi - FIGURE 1B TEST PERIOD 2 Performance summary for NCBS-2, IBM five-min. bars. Feb. 28 to March 30. Statistics based upon buying and selling 1,000 shares of IBM. Performance summary: All trades Total net profit (): Gross profit (): Total no. of trades: Number winning trades: Largest winning trade (): Average winning trade (): Ratio avg. win/avg. loss: Max. consec. winners: Avg. Nein. bars in winners: Max. intraday drawdown (): Profit factor: 9,640 34,460 38 20 5,350 1,723 1.25 3 48 -10,030 1.39 Max. Nein. contracts held: 1 Open position P/L (): Gross loss (): Percent profitable (): Number losing trades: Largest losing trade (): Average losing trade (): Avg. trade(win loss) (): Max. consec. losers: Avg. Nein. bars in losers: 0 -24,820 52.63 18 -3,400 -1,378.89 253.68 2 28 Source: TradeStation by TradeStation Group Inc. 44 activetradermag October 2001 ACTIVE TRADER tional information on optimization and walk-forward testing. Of the four system parameters to find (nhi, nlo, xoU and xoD), the best parameters are defined as those values that give the best net profits and best total winning bars/total losing bars ratio with the minimum drawdown and minimum largest losing trades. In addition, the results should be stable e. g. the profits, wins and drawdowns should not change by much as the parameters move by a small amount away from their optimum values. Also, in choosing the best parameters, we considered only those parameter sets with maximum consecutive losses of four or less. Improved performance The optimum parameters in Table 1 show the first test data section produced the same optimum parameters as the original NCBS. This can been seen by observing that both xoU and xoD are exactly the same and are equal to f of the original NCBS. The sample performance summaries in Figures 1a and 1b, and the out-of-sample performance summary of Figure 2a, show the out-of-sample performance was better than the test sample performance with respect to average winning and losing trades, drawdowns and profit factor. This improved performance in the out-of-sample section could have been due to Results Table 1 shows the optimum parameters for the IBM five-minute data series. The lookback periods refer to number of bars and the filters values are given as dollar amounts. Figures 1a and 1b) show the performance summary of the test windows using the optimum parameters shown in Table 1. Figure 2a shows the combined performance summary of the two out-ofsample data segments from March 26 to April 6 for NCBS-2. This performance represents what would have happened in real time if the parameters found in the test sections (Table 1) were used. Slippage and commissions are not included. For comparison, Figure 2b (bottom, right) shows the combined performance summary of the two out-ofsample data segments from March 26 to April 6 for the original NCBS. Figure 3 shows a specialized percentage trade-by-trade summary from March 26 to April 6. Note that the trades from March 26 to April 6 are the out-of-sample trades generated from the optimized parameters from the two test sections of Feb. 21 to March 23 and Feb. 28 to March 30. The in-sample trades are, by definition, curve-fit and are not of interest here. In addition, for comparison with Figure 3, Figure 4 contains the specialized trade-by-trade summary from the original NCBS for the same out-of-sample dates. Figure 5 is a five-minute chart of IBM with the NCBS-2 channels superimposed and some of the buy and sell signals from the Figure 3 trade-by trade summary indicated on the charts. (All the signals, as well as expanded performance statistics, can be found at activetradermag.) Also included at the bottom of the chart is the bar-by-bar profit or loss of each trade. FIGURE 2A TEST PERIOD 1 Combined walk-forward out-of-sample performance summary for NCBS-2, IBM five-min. bars, March 26 to April 6. Statistics based upon buying and selling 1,000 shares of IBM. Performance summary: Total net profit (): Gross profit (): Total no. of trades: Number winning trades: Largest winning trade (): Average winning trade (): Ratio avg. win/avg. loss: Max. consec. winners: Avg. Nein. bars in winners: Max intraday drawdown (): Profit factor: 8,650 15,390 15 7 4,000 2,198.57 2.61 2 57 -5,660 2.28 Max. Nein. contracts held: 1 All trades Open position P/L (): Gross loss (): Percent profitable (): Number losing trades: Largest losing trade (): Average losing trade (): Avg. trade(win loss) (): Max. consec. losers: Avg. Nein. bars in losers: 0 -6,740 0.47 8 -1,730 -842.50 576.67 3 38 FIGURE 2B TEST PERIOD 2 Combined walk-forward out-of-sample performance summary for NCBS, IBM five-min. bars, March 26 to April 6. Statistics based upon buying and selling 1,000 shares of IBM. Performance summary: Total net profit (): Gross profit (): Total no. of trades: Number winning trades: Largest winning trade (): Average winning trade (): Ratio avg. win/avg. loss: Max. consec. winners: Avg. Nein. bars in winners: Max. intraday drawdown (): Profit factor: 8,390 14,460 16 8 4,000 1,807.50 2.382 5 54 -4,480 2.382 Max. Nein. contracts held: 1 All trades Open position P/L (): Gross loss (): Percent profitable (): Number losing trades: Largest losing trade (): Average losing trade (): Avg. trade(win loss) (): Max. consec. losers: Avg. Nein. bars in losers: 0 -6,070 50 8 -1,350 -758.75 524.375 3 37 Source: TradeStation by TradeStation Group Inc. ACTIVE TRADER October 2001 activetradermag 50 FIGURE 3 OUT-OF-SAMPLE TRADE BY TRADE SUMMARY IBM five-min. NCBS-2 Trade size 1,000 shares March 26 to April 6 Entry date Entry time Sell Buy Sell Buy Sell Buy Sell Sell Sell Buy Sell Buy Sell Buy Sell Entry price 93.75 95.59 97.92 96.05 94.90 96.70 96.20 96.45 93.33 92.99 92.55 95.68 97.30 98.75 97.02 Exit date 3/26/01 3/27/01 3/28/01 3/29/01 3/29/01 3/30/01 3/30/01 4/2/01 4/3/01 4/4/01 4/4/01 4/5/01 4/6/01 4/6/01 4/6/01 Exit time 15:55 15:55 15:55 15:05 15:55 13:05 15:55 15:55 15:55 12:55 15:55 15:55 12:00 12:35 15:55 Exit Bars Trade price in trade PL 94.52 99.59 94.50 94.90 94.88 96.20 96.25 94.50 90.50 92.55 91.85 98.15 98.75 97.02 97.67 67 68 75 60 10 41 34 60 75 24 36 75 28 7 40 (770) 4,000 3,420 (1,150) 20 (500) (50) 1,950 2,830 (440) 700 2,470 (1,450) (1,730) (650) Trade PL (0.82) 4.18 3.49 (1.20) 0.02 (0.52) (0.05) 2.02 3.03 (0.47) 0.76 2.58 (1.49) (1.75) (0.67) Trade Max Pft 0 4,300 3,420 950 390 800 220 2,650 3,070 910 930 3,040 550 1,150 20 Time Trade Max DD 0 (380) (500) (840) (670) (660) (520) (10) Time 3/26/01 10:20 3/27/01 10:15 3/28/01 9:40 3/29/01 10:05 3/29/01 15:05 3/30/01 4/2/01 4/3/01 4/4/01 4/4/01 4/5/01 4/6/01 4/6/01 4/6/01 9:40 10:55 9:40 10:55 12:55 9:40 9:40 12:00 12:35 3/30/01 13:05 10:20 15:50 12:00 10:30 15:15 11:55 13:15 15:40 15:40 11:20 14:00 15:25 11:15 12:05 12:35 (1,620) 10:35 10:15 9:40 15:45 14:35 9:45 11:00 13:20 9:40 (1,160) 11:20 (1,190) 10:00 (1,250) 11:40 (1,450) 12:00 (1,730) 12:35 (2,240) 13:55 Average (948) Total Average Average 8,650 0.61 1,493 Source: Meyers Analytics FIGURE 4 OUT-OF-SAMPLE TRADE BY TRADE SUMMARY IBM five-min. NCBS Trade size 1,000 shares March 26 to April 6 Entry date 3/26/01 3/27/01 3/28/01 3/29/01 3/29/01 3/30/01 3/30/01 4/2/01 4/2/01 4/3/01 4/4/01 4/4/01 4/5/01 4/6/01 4/6/01 4/6/01 Entry time 10:20 Sell 10:15 Buy 9:40 Sell 10:05 Buy 15:05 Sell 9:40 9:40 Buy Buy 13:05 Sell 10:55 Sell 10:00 Sell 9:45 9:40 9:40 Buy Buy Sell 13:50 Sell Entry price 93.75 95.59 97.92 96.05 94.90 96.70 96.20 97.75 96.40 93.00 92.00 92.00 95.68 97.30 98.24 97.30 Exit date 3/26/01 3/27/01 3/28/01 3/29/01 3/29/01 3/30/01 3/30/01 4/2/01 4/2/01 4/3/01 4/4/01 4/4/01 4/5/01 4/6/01 4/6/01 4/6/01 Exit time 15:55 15:55 15:55 15:05 15:55 13:05 15:55 10:55 15:55 15:55 13:50 15:55 15:55 11:55 12:35 15:55 Exit Bars Trade price in trade PL 94.52 99.59 94.50 94.90 94.88 96.20 96.25 96.40 94.50 90.50 92.00 91.85 98.15 98.24 97.30 97.67 67 68 75 60 10 41 34 15 60 71 49 25 75 27 8 40 (770) 4,000 3,420 20 (500) (50) (1,350) 1,900 2,500 0 150 2,470 (940) (940) (370) Trade PL (0.82) 4.18 3.49 0.02 (0.52) -0.05) -1.38) 1.97 2.69 0.00 0.16 2.58 (0.97) (0.96) (0.38) Trade Max Pft 0 4,300 3,420 950 390 800 220 350 2,600 2,740 1,900 380 3,040 550 1,660 300 Time Trade Max DD 0 (380) (500) (840) Time 10:20 15:50 12:00 10:30 15:15 11:55 13:15 10:05 15:40 15:40 11:20 14:00 15:25 11:15 12:05 12:35 (1,620) 10:35 10:15 9:40 15:45 14:35 (1,150) (1.20) (1,160) 11:20 (1,190) 10:00 (1,350) 10:55 (1,300) 11:40 0 (500) (10) (940) (940) Average (911) 10:00 14:20 9:40 11:55 12:35 (1,890) 10:30 11:55 Buy 12:35 Sell (1,960) 13:55 Total Average Average 8,390 0.55 1,475 Source: Meyers Analytics 51 activetradermag October 2001 ACTIVE TRADER FIGURE 5 NCBS-2 SIGNALS Trade signals for the NCBS-2 are shown on a five-minute chart of IBM. The blue and red lines are the long and short filter levels, respectively. International Business Machine (IBM), five-minute 0 100 101 -1 99 98 97 96 -1 1 0 0 95 94 93 4,000 2,500 1,000 500 9:55 10:50 11:45 12:40 1:35 2:30 3/27 10:50 11:45 12:40 1:35 2:30 3/28 10:50 11:45 12:40 1:35 2:30 chance but does indicate that four weeks of test data were enough to capture the intraday price dynamics of IBM. The performance summaries in Figures 2a and 2b show there is very little difference between the NCBS and NCBS-2. The less-complicated NCBS, while having a slightly lower net profit and average win/average loss ratio, has a smaller drawdown and a smaller largest losing trade. Comparison of Figures 2a and 2b favors the simpler NCBS. The out-of-sample trade-by-trade summary of Figure 3 shows the system did better on short trades than on long trades. This could indicate a negative bias for the system, or perhaps, given the current bear market, this could be normal. Whatever the reason, this bias warrants further investigation. There were no big winners or big losers, indicating steady returns. Average wins were 2.6 times average losses in the out-ofsample section. Average trade run-ups were 1,493, average trade drawdowns were -948 and the average trade net profit was 576. Its also instructive to compare Figure 3 with Figure 4 to ACTIVE TRADER October 2001 activetradermag determine if the more complicated NCBS-2 offers any advantage in the trade-by-trade figures. There seems to be little advantage: Both systems totals and averages are nearly the same. The NCBS-2 had one less trade and slightly better numbers. However, the difference wasnt enough to claim any superiority or to justify the added complication of another optimization parameter. The NCBS-2 did very well in catching every major intraday trend of IBM. The charts show the system constraint of not carrying positions overnight eliminated many negative opening surprises. Overall, the system did a good job in minimizing the losses resulting from the inevitable whipsaws that will occur in any trading system and maximizing the profits from the major intraday trend moves of IBM. To use NCBS-2 in real time trading, the results from at least 10 to 20 more tests and out-of-sample periods would have to be examined to make sure that the results above were not due to pure chance. 52 ADVANCED Strategies The multibar range BREAKOUT SYSTEM Breakouts of price channels can be profitable if the volatility is there and youre on the right side of the trade. This stop-and-reverse system tries to capture intraday trends in the SP E-Mini contract by recognizing differences in the characteristics of up moves and down moves. BY DENNIS MEYERS, PH. D. FIGURE 1 TRADESTATION CODE FOR THE MULTIBAR RANGE BREAKOUT SYSTEM Input: n(45),bx(0.45),m(15),sx(0.45),XTime(1515) vars: hhv1(h),llv1(l),hhv2(h),llv2(l),ii(0),xb(c),xs(c) hhv1h llv1l for ii1 to n-1 begin if hiihhv1 then hhv1hii if liihhv2 then hhv2hii if lii
No comments:
Post a Comment