Robert Fischer - Fibonacci Applications & Strategies for Traders

8/6/2019 Robert Fischer - Fibonacci Applications & Strategies for Traders

1/94


2/94

WILEY TRADER'S ADVANTAGE SERIESJohn F. Ehlers, MESA and Trading Market CyclesRobert Pardo, Design, Testing, and Opt imiza tion of Trading SystemsGary Klopfenstein and Jon Stein, Trading Currency Cross RatesRobert Fischer, Fibonacci Applications and Strategies for Traders

,I

F i b o n a c c i R p p l i c a t i o n sa n d S t r a t e g i e sf o r T r a d e r s

H o b e r t F i s c h e rSeries Editor: Perry 1. KaufmanJ o h n Wiley & Sons, Inc.

New York Chichester Brisbane Toronto Singapore


3/94

--

This text is printed on acid-free paper.Copyright io 1993 byJohn Wiley& Sons, Inc.All rights reserved. Published simultaneously in Canada.Reproduction or translat ion ofany part ofthis work beyondtbat permitted by Section 107 or 108 ofthe 1976UnitedStates Copyright Act without the permission ofthe copyrightowner is unlawful, Requests forpermission or furtherinformation should be addressed to the Permissions Department,John Wiley& Sons, Inc., 605 Third Avenue, NewYork,NY10158-0012.

To Jens and Claudius,who graciously tolerated my absences,and who weremissed over the many yearsmore than I can ever say.

This publication is designed to provide accurate andauthuritative information in regard to the subjectmatter covered. It is soldwith the understanding thatthe publisher is nut engaged in rendering legal, accounting,or other professional services. If legal advice or otherexpert assistance is required, the services ofa competentprofessional person should be Bought From a Declarationof Pri nciplas jointly adopted bya Committee oftheAmerican Bar Association and a Committee ofPublishers.Library ofCongress Cataloging in Publication Data:Fischer, Robert, 1941 .Iune 17-Fibonacci applications and strategies for traders .I byRobertFischer.

p. em. - (Wiley trader's library)Includes index.ISBN 0-471-58520-31. Speculation. 2. Stocks. 3. Commodity exchanges. 4. Fibonaccinumbers. 1. Title. II. SeriesHG6041.F573 1993332.64-dc20 93-7637Printed in the United States ofAmerica10 98 76 54 3 2 1


4/94

ACKNOWLEDGMENTS

In 1977Mr. Joachim Bahr-Vollrath ofGermany introduced me to theprinciples of Elliott and Fibonacci. Without the generous sharing ofhis knowledge-the foundation for this book, I could never have pro-ceeded. Thank you,Mr. Bahr!

After my1983Fibonacci seminars, Iwas fortunate tomeet HomerRussell, Homerdid an incredible jobcreating the software onthe loga-rithmic spiral and time analysis used in this book.Without Homer'sprogramming skills, mostofthe substantiation ofthe theory would nothave been possible.

Even though the knowledge and the programs have been in placesince 1985, it took until 1992for me to put the loose ends together. Itstarted with a visit to Omega Research, Inc. in Miami. Bill Cruz gaveme the necessary support to print most ofthe charts with TradeSta-tion, one of the leading on-line technical information systems. Moreimportantly, I got to know Samuel Tennis. Sam is a senior softwaretechnician with Omega Research. His programming skills made itpossible for me to analyze corrections and extensions in a way neverbefore possible. Sam's personal engagement contributed substantiallyto this book.

Mybiggest problemwas getting mentally organized to write this.My son,-Iens,a student at the University ofCologne,Germany, was of

v ii


5/94

viii ACKNOWLEDGMENTS

invaluable help in getting me started and guiding methrough the con-cepts fromonechapter to the next. Workingso closelywith myson wastruly the great reward of this book.

Germany is far away and writing in a second language is any-thing but easy. Whenever I had a chapter drafted, I gave it to AnneCavanagh. With her indepth knowledge ofthe markets and her writ-ing skills, she helped to get this bookfinished.Mr. Perry Kaufman and I havecomea longwaytogether. Wecol-laborated on my first work in 1979, but this time Perry gave me somuch more. It is his flair for the written wordthat shapes this book-not to mention his patience and continued drive for perfection (whichhurt at times, but made me work even harder). Thank you, Perry!

R.F.

PREFACE

This book deals with the most fascinating subject in stock and com-modity analysis. It is the integration ofNature's Law(seen as a behav-ioral phenomenon) into price studies to create a serious and reliableinvestment tool.Newdiscoveries with the Fibonacci summation seriesopenthe window for a price-time analysis that can pinpoint tops andbottoms with a stunning accuracy.

For 50 years, R.N. Elliott has remained a legend for legions ofanalysts. Without his ideas, this bookcould never have been written.His brilliant deductions include: Market swings are a reflection ofhuman behavior. Human behavior can be related to a phenomenon ofnature. Nature's law can bemeasured byusing the Fibonacci summationseries.From his findings, published in The Wave Principles, Elliott

claims that he can forecast price movements.Here weseparate ourselves fromElliott. Weconsider this impos-

sible and will prove it. Because ofthe wave count, the Elliott conceptbecomeshighly subjective.

ix


6/94

x PREFACE

Instead, we approach the markets by focusing strictly on theFibonacci summation series and ignoring the wave count. This re-moves all subjectivity and replaces the uncertainty with definitive,tested rules. The introduction of entry and exit rules makes disci-plined, even automatic trading possible.

The bookbegins with an indepth study ofthe Fibonacci summa-tion series, emphasizing the Fibonacci ratio. It is always surprising tolearn that major achievements in the fields ofnatural science, nucleartheory, radio, and television have this ratio in common.The Fibonacciratio is not just a numbers gamebut one ofthe most important mathe-matical presentations ofnatural phenomena ever discovered.Weapplythis ratio as a geometric tool to equity and commodity price swingsusing techniques never seen before.

The Fibonacci ratio is best used in forecasting correction targets.However, different strategies are presented, including a very short-term approach for those investors who donot want to hold a positionlonger than one ortwo days. Acomputer study was used to confirm thevalidity ofthis strategy.

In addition, the Fibonacci ratio is used to analyze price targetson extensions. Extensions happen in run-away markets. An invest-ment entered at the end ofthe extension isconsidered the safest strat-egy in the Elliott concept.

I introduced the idea of time analysis in 1983 at a seminar inChicago. Nine years later, with the aid of a computer, it is nowpossi-ble to showthat this analysis is a valid investment alternative.

The final chapter introduces the logarithmic spiral, the invest-ment tool that takes a lot ofimagination to believe in. Even though wenow have a computer program to draw the spirals, it took us manyyears to understand it and to developrules that make it easy to use forinvestment decisions. This spiral is the link between Nature's Lawand human behavior, expressed in the price pattern ofstocks and com-modities. Price patterns don't happen by accident. The logarithmicspiral allowsus to analyze market swings both "price and time" witha precision never seen before.

This bookis intended to be educational. Within the size limits ofthis book, the concepts are presented thoroughly with detailed exam-ples. I hope that you find these ideas as exciting, enlightening, anduseful as I have.

ROBERT FISCHERChicago, Illinois

--

THE TRADER'S ADVANTAGESERIES PREFACE

The Trader's Advantage Series is a newconceptin publishing fortradersand analysts of futures, options, equity, and generally all world eco-nomicmarkets. Books in the series present single ideas with only thatbackground information needed to understand the content. Nolong in-troductions, no definitions ofthe futures contract, clearing house, andorder entry. Focused.

The futures and options industry is no longer in its infancy.From its role as an agricultural vehicle it has become the alterego ofthe most active world markets. The use ofEFPs (exchange for physi-cals) in currency markets makes the selection ofphysical or futuresmarkets transparent, in the same way the futures markets evolvedinto the official pricing vehicle for world grain. With a single tele-phone call, a trader or investment manager can hedge a stock portfolio,set a crossrate, perform a swap, or buy the protection ofan inflationindex. The classic regimes can no longer be clearly separated.

And this isjust the beginning. Automated exchanges are pene-trating traditional openoutcry markets. Even now,from the time thetransaction is completed in the pit, everything else is electronic."Program trading" is the automated response to the analysis of a

xi


7/94

xii T HE T RA DE R'S A DV AN TA GE S ER IE S P RE FA CE

computerized ticker tape, and it isjust the tip ofthe inevitable evolu-tionary process. Soon the executions will be computerized and thenwe won't be able to call anyone to complain about a fill. Perhaps wewon't even have to place an order to get a fill.

Market literature has also evolved.Many ofthe books written ontrading are introductory. Even those intended formore advanced audi-ences often include a review of contract specifications and marketmechanics. There are very fewbooks specifically targeted for the ex-perienced and professional traders and analysts. The Trader's Advan-tage Series changes all that.

This series presents contributions by established professionalsand exceptional research analysts. The authors' highly specialized tal-ents havebeen applied primarily to futures, cash, and equity marketsbut are often generally applicable to price forecasting. Topicsin theseries will include trading systems and individual techniques, but allare a necessary part ofthe developmentprocess that isintrinsic to im-proving price forecasting and trading.

These works are creative, often state-of-the-art. They offer newtechniques, in-depth analysis of current trading methods, or innova-tive and enlightening waysof looking at still unsolved problems. Theideas are explained in a clear, straightforward manner with frequentexamples and illustrations. Because they donot contain unnecessarybackground material they are short and to the point. They requirecareful reading, study, and consideration. In exchange, they con-tribute knowledge to help build an unparalleled understanding of allareas ofmarket analysis and forecasting.

I first met Robert Fischer through his manuscript, Stocks orOp-tions? Programs for Profits (Wiley, 1980). It has originally been pub-lished in German and he was struggling to make the concepts clear toEnglish-speaking readers. Iwasimpressed byhis intense desire tocom-municate that information to others. I helped him with the translationbecause of my belief that every author wants the quality of his or herpublication to be the very best.

Robert has always been fascinated bythe patterns that continu-ally appear on price charts. Longbefore his options book, he had de-voured all the information available on Elliott's waveprinciples. Hetraded Elliott's method for years, always looking for ways to improveits consistency and remove subjectivity.

When he began to realize the importance ofFibonacci numberswithin chart analysis, you could see the electricity in his research. He

TH E TR AD ER 'S A DV AN TA GE SE RIES PRE FA CE xiii

absorbed the subtleties ofFibonacci ratios, studying pl.ants, a~chite~-ture, and human behavior. In 1983, I at~ended the FIbonaCCISemi-nars where Robert introduced the use oftime goal days as a method offorecasting based onFibonacci ratios. Everyone was enthr~lled byt~econcept that human behavior often favored certain proportions. Whilemost traders nowembrace Fibonacci retracement levelsof 38 percent,50 percent, and 62 percent, Robert was still far ahead of them by ap-plying these ratios to forecasting event days. He added depth and con-sistency to chart trading.But he was still unhappy with the results. I watched Robertsearch to find a solution that was more accurate and yet consistentwith what he believesis the underlying behavioral force in the market.Then he found the logarithmic spiral, one of the most sophisticatedphenomenon in nature. Integrated into his analysis, it combines bothprice and time in a dynamic way.In the mid-1980s, I was asked to program the formulas so thatthese logarithmic spirals could appear on a screen along with theprice chart. A trader would be able to sit and watch price m?vement.asit reached a support or resistance level indicated bythe spiral, Be~nggoodat math, I underestimated the difficulties and was soon s.wIm-ming in pages offormulas. And, this was before I got to the pom~ofdeciding whether to rotate the spiral clockwise or counterclockwIse.

Fortunately, it wasjust then that Robert found HomerRussell. Imade a graceful exit, but was disappointed that Homer d~dn't.see~ toneed any ofmy equations. It wasn't longbefore the logarithmic spiralprogram was operational.In this book, Robert Fischer coversall ofthe important steps ~hatresulted in combining Elliott and Fibonacci in his unique way,bring-ing it to the point ofits development today. It is filled with new ~deasand clear expectations. Until you reach the very end, the techmquesrequire nomathematical skills. After that, a computer program pro-vided in Appendix Bwill dothe rest.I am very pleased that Robert has been able to write this book. Iknow ofno one with a better understanding of Elliott's method andFibonacci's principles, and noonewhohas workedharder to put themtogether.

PERRY J_ KAUFMANVermontJune 1993


8/94I ".- .. '': ,

CONTENTS

CHAPTER 1 THE FIBONACCI RATIOThe Fibonacci Summation SeriesThe Divine Proportion in NatureThe Fibonacci Ratio in Geometry

1136

CHAPTER 2 THE ELLIOTT WAVE PRINCIPLEIN A NUTSHELL

Elliott's Market PatternsFibonacci RatioTrend ChannelsSummary

1114182021

CHAPTER 3 WORKING WITH A 5-WAVE PATTERNPredicting the End ofWave5 Using a Trend ChannelPredicting the End ofWave5 with Fibonacci RatiosThe Amplitude ofWaves 1, 2, and 3, and the FibonacciRatio 0.618Investing with OptionsSummary

252629334042

CHAPTER 4 WORKING WITH CORRECTIONSRules That Are ReliableWhen Not to InvestSize ofCorrections

45464950

xv


9/94

xv i CONTENTS

Corrections in a Long-Term TrendCorrections in a Short-Term TrendBigCorrections and Trend ChangesUse ofthe Options MarketSummary

5354687070737687939395

9510 010 210 310 410 611312 112 212 412 712 713 013 213 313 413 815 216 116 3

16 5

16 7

.,,

CHAPTER 5 WORKING WITH EXTENSIONSExtensions in Wave3Extensions in Wave5Using the Options MarketSummary

CHAPTER 6 MULTIPLE FIBONACCI PRICE TARGETSCombining Daily 5-WavePatterns andWeeklyCorrectionsCombining Extensions and CorrectionsSummary

CHAPTER 7 TIME ANALYSISTime Goal DaysTrading Using Time AnalysisMoreon the Structure ofTime Goal DaysReviewAdditional RulesSummary

CHAPTER B COMBINING PRICE AND TIMEConcept ofCombining Price and TimeABritish Pound ExampleSummary

CHAPTER 9 THE LoGARITHMIC SPIRALThe Concept ofthe SpiralMore on the Structure ofthe SpiralWorkingwith the SpiralSummary

APPENDIX A THE GoLDEN SECTION COMPASS

APPENDIX B SPIRAL EQUATION AND COMPUTERPRoGRAM

INDEX

F i b o n a c c i R p p l i c a t i o n sa n d s t r a t e g i e sf o r T r a d e r s


10/94

1THE FIBONACCI RATIO

TH E FIB ONACCI SU MM ATION SER IESLet your imagination soar. Think of the universe, the constellations,the galaxy. Contemplate the beauty and formof all the wonders ofna-ture: the trees, the oceans, flowers, plant life, animals, and even themicroorganism in the air we breathe. Give further thought to theachievements ofman in the fields ofnatural science, nuclear theory,medicine, radio, and television. Itmaysurprise you to learn that all ofthese haveone thing in common-the Fibonacci Summation Series.

In the thirteenth century, Thomas ofAquinas described one ofthe basic rules ofaesthetics-man's senses enjoyobjects that are prop-erly proportioned. He referred to the direct relationship betweenbeauty and mathematics, which is often measurable and can be foundin nature. Man instinctively reacts positively to clear geometricalforms, in both his natural environment and in objects created byhim,such as paintings. Thomas ofAquinas was referring to the same prin-ciple that Fibonacci discovered.

Fibonacci, a mathematician, lived in 1175. He was one of themost illustrious scientists of his time. Among his greatest achieve-ments was the introduction ofArabic numerals to supersede the Ro-man figures. He developedthe Fibonacci summation series:

1,1,2,3,5,8,13,21,34,55,89,144, 1


11/94

2 THE FIBONACCI RATIO THE DIVINE PROPORTION IN NATURE 3

This mathematical series develops when, beginning with 1,1,the next number is formed from the sum ofthe previous two numbers.But what makes this series so important?

The series tends asymptotically (approaching slower and slower)toward a constant ratio. However,this ratio is irrational, that is, it hasa never-ending, unpredictable sequence of decimal values stringingafter it. It can never be expressed exactly. Ifeach number in the seriesis divided by its preceding value (e.g., 13 -;-8), the resul t is a ratio thatoscillates around the irrational 1.61803398875 ... , being higherone time and lowerthe next. But never in eternity can the precise ratiobe known to the last digit. For the sake of brevity, we will refer to itas 1.618.

This ratio had begun to gather special names even before LucaPacioli (a medieval mathematician) named it the Divine Proportion.Among its current day names are the Golden Section, the GoldenMean, and the Ratio of Whirling Squares. Kepler called the ratio "oneofthe jewels in geometry." Algebraically it is generally designated bythe Greek letter phi ($= 1.618).

The asymptotic tendency of the series, its ratio's ever-tighteningoscillation around the irrational phi, can be best understood by show-ing the ratios of the first few entries in the series. This example takesthe ratio of the second entry to the first, the third to the second, theforth to the third, and so forth:

is simply the reciprocal of 1.618 (1 -i- 1.618). But this is also a very un-usual, even remarkable phenomenon. Since the original ratio has noend, this ratio must also have no end.

Another important fact is that the square ofany Fibonacci num-ber is equal to the number in the series before it, multiplied by thenumber after it, plus or minus 1.

52 = (3 x 8) + 182 = (5 x 13) - 1132 = (8 X 21) + 1

Plus and minus continually alternate. Again this phenomenon is an im-plicit part ofthe Elliott WavePrinciple called the Rule ofAlternation. Itstates that complex corrective waves alternate with simple ones, strongimpulse waves with weak impulse waves, and soon.

TH E D IV INE PROPORTION IN N ATURE

1: 1 = 1.0000, which is lower than phi by 0.61802: 1 = 2.0000, which is higher than phi by 0.38203: 2 = 1.5000, which is lowerthan phi by 0.11805: 3 = 1.6667, which is higher than phi by 0.04868: 5=1.6000, which is lowerthan phi by 0.0180

It is remarkable howmany constant values can be calculated using theFibonacci sequence, and howthe separate figures ofthe sequence recurin somany variations. However,it cannot be stressed too strongly thatthis is not just a numbers game but the most important mathematicalpresentation of natural phenomena ever discovered. The followingillus-trations depict some interesting applications of this mathematicalsequence.

As we continue in the Fibonacci summation series, each memberwill divide into the next one by a closer and closer approximation ofphi which can never be reached.

Wewill see later that the individual numbers in the Fibonaccisummation series can be seen in commodity price movements. Theswings of the ratios around the value 1.618 by either higher or lowernumbers will be found in the Elliot Wave Principle described as theRule of Alternation. Man subconsciously seeks the Divine Proportion;it satisfies his comfort level.

Dividing any number ofthe Fibonacci sequence bythe followingnumber in the series asymptotically approaches the ratio 0.618, which

The Pyramid ofGizehMany people have tried to penetrate the secrets of the Pyramid ofGizeh. It is different from the other Egyptian pyramids because it isnot a tomb, but rather an unsolvable puzzle of figure combinations.The marvelous ingenuity, skill, time, and labor used bythe designersofthe pyramid to erect a perpetual symbol demonstrates the supremeimportance of the message they desired to convey to posterity. Thatera was preliterary and prehieroglyphic and symbols were the onlymeans of recording inventions.

The key to the geometrical and mathematical secret ofthe Pyra-mid of Gizeh, so long a puzzle to mankind, was actually handed toHerodotus by the temple priests when they informed him that thepyramid was designed in such way that the area of each of its faceswas equal to the square ofits height (Figure 1-1).


12/94

4 THE FIBONACCI RATIO THE DIVINE PROPORTION IN NATURE 5

Araa 01 square280 x 280 ~ 78,400

FI~T bC lS ,lobO # . ~ $ ~ :; tSTl


13/94

6 THE FIBONACCI RATIO

Figure 1-3 Fibonacci numbers found inthe f lowers ofsneezewort. (Source: TheDivine Proportion, by H.E. Huntley (New York: Dover, 1970) p. 163. Reprintedwith permission.)

IrisPrimroseRagwortDaisyMichalmas Daisy

3 petals5 petals13petals34 petals55 and 89 petals

The number and arrangement ofthe florets in the head ofamem-ber of the composite family is a particularly beautiful example ofgoldennumbers found in nature.

Wewere looking for laws that workedin the past and thereforehave the highest probability ofcontinuing toworkin the future. In theFibonacci ratio, we seemto have found such a Jaw.

THE FIBONACCI RATIO IN GEOMETRYThe existence ofthe Fibonacci ratio in geometry is well known, butapplying this ratio as a geometric tool to commodity pric~swith .sp~-rals and ellipses has never been done before. The reason ISthat It ISnecessary to use the powerof computers to apply the logarithmic spi-ral and logarithmic ellipse as an analytic tool.

THE FIBONACCI RATIO IN GEOMETRY 7

Both the spiral and ellipse have unusual properties consistentwith the Fibonacci ratio in two dimensions, price and time. It is verylikelythat the integration ofspirals and ellipses will elevate the inter-pretation and use ofthe Fibonacci ratio to another, much higher level.Up to now the Fibonacci ratio was used for measurement of correc-tions and extensions ofprice swings. The time element forecast wasseldomintegrated because it did not seemtobe as reliable as the priceanalysis. By including spirals and ellipses in a geometric analysis,both price and time analysis can be combinedaccurately.

The Golden Section of a LineThe Greek mathematician Euclid related the Golden Section to astraight line (Figure 1-4). The lineAB of'length L is divided into twosegments bypoint C. Let the length ofAC and CB be a and b, respec-tively. If C is a point such that L :a equals a :b, then C is the GoldenSection AB. The ratio L:a ora: b is called the "Goldenratio." In otherwords, the point C divides the line AB into two parts in such a waythat the ratios of those parts is 1.618and .618.

A c B----- L ----+

Golden cuiFigure 1-4 Golden section of a line.

The Golden Section of a RectangleIn the Great Pyramid, the rectangular floor ofthe King's chamber il-lustrates the GoldenSection (Figure 1-5). A "GoldenRectangle" canbest bedemonstrated bystarting with a square-the basic area ofthePyramid ofGizeh. SideAB ofthe square ABCD in Figure 1-5 is bi-sected. With the center E and radius EC, an arc ofa circle is drawncutting the extension ofAB at F Line FG is drawn perpendicular toAF, meeting the extension of DC at G. Then AFGD is the GoldenRectangle. According to the definition, the Golden Section Rectangleis 1.618 times longer than it is wide. The ratio of its proportions istherefore phi:

1.618:1

J


14/94

8 T H E FIBONACCI RATIO

A~ ~E~ ~Br- ~F

,,o c G

Figure 1-5 Golden sect ion of a rectangle.

Greek architects and sculptors incorporated this ratio into theirwork. Phidias, a famous Greek sculptor, made use of it; the propor-tions of the Parthenon at Athens illustrate this. Built in the fifth cen-tury B.C., it is based on a triangular pediment, which is still intact. Itsdimensions exactly fit into the Golden Rectangle. It stands as an ex-ample of the aesthetic value of this unique shape.

The Logarithmic SpiralThe only mathematical curve to follow the pattern of growth is the log-arithmic spiral, expressed in the spira mirabilis or the nautilus shell(Figure 1-6). The logarithmic spiral is called the most beautiful ofmathematical curves. This spiral has been a common occurrence inthe natural world for millions of years. The Golden Section and theFibonacci series are all associated with this remarkable curve.

Figure 1-6 shows a radiograph of the shell ofthe chambered nau-tilus (nautilus pompil ius). The successive chambers of the nautilusare built on the framework of a logarithmic spiral. As the shell grows,the size ofthe chambers increases, but their shape remains unaltered.

We will demonstrate the geometry of the logarithmic spiral us-ing the Golden Rectangle ABCD (Figure 1-7) with AB :BC = phi:l.Through point E, called the "Golden Cut" ofAB, line EF is draw n per-pendicular to AB cutting the square AEFD from the rectangle.The remaining rectangle EBCF is a Golden Rectangle. Ifthe squareEBGH is lopped off, the remaining figure HGCF is also a GoldenRectangle. Assume that this process is repeated indefinitely untilthe limiting rectangle 0 is so small that it is indistinguishable froma point.

T H E FIBONACCI RATIO IN GEOMETRY 9

Figure 1-6 Logarit hmic spiral. (Source: TheDivine Proportion, by H.E. Huntley(New York: Dover, 1970) p. iv, Reprinted with permission.)

The limiting point 0 is called the pole of the equal angle spiral,which passes through the Golden Cuts D, E, G, J . . . (the sides ofthe rectangle are nearly, but not completely, tangential to the curve).

The relation of the Fibonacci series is evident from Figure 1-7,for the spiral passes diagonally through opposite corners of successivesquares, such as DE, EG, GJ, .... The lengths of the sides of thesesquares form a Fibonacci series. If the smallest square has a side oflength d, the adjacent square must also have a side of length d. The

Figure 1-7 Geometry of the logarithmic spiral. (Source: TheDivine Proportion,by H.E. Huntley (New York: Dover, 1970) p. 101. Reprinted with permission.)


15/94

10 THE FIBONACCI RATIO

Figure1-8 Logarithmic ellipse.(Source; TheDivine Proportion, by H.E. Huntley(New York: Dover, 1970) p. 71. Reprinted with permission.)

next square has a side oflength 2d (twice as long),the next of3d, andsoon, forming the series Id, Id, 2d, 3d, 5d, 8d, 13d, ... which is ex-actly the Fibonacci Series 1,1,2,3,5,8, 13, ....

Two segments ofthe spiral maybe different in size, but they arenot different in shape. The spiral is without a terminal point; whilegrowing outwards (or inwards) indefinitely, its shape remains un-changed. The logarithmic spiral is the link between the FibonacciSummation Series and the WorldofNature.

Th e Logarithmic EllipseImportant curves are found in nature. The most significant to civi-lization include the horizon of the ocean, the meteor track, theparabola ofa waterfall, the arcs traced in the sky by the sun and thecrescent moon, and the flight ofa bird.

An ellipse is the mathematical term for an oval. Each ellipse canbe precisely designated by only a few characteristics. The extremeform of an ellipse is the parabola (Figure 1-8), which can be repre-sented mathematically as:

y2 = 4axPoint P is equidistant from a fixed point F (the focus)and a fixed

line ZM (the directrix). The curve is symmetrical about the axis.

2THE ELLIOTT WAVEPRINCIPLE IN A NUTSHELL

Ralph Nelson Elliott was an engineer. After a serious illness in theearly 1930s, he turned to the analysis of stock prices, especially theDowJones Index. After a number ofremarkably successful forecasts,Elliott published a series ofarticles in Financial World Magazine in1939. In these, he first presented his contention that the DowJonesIndex movesin rhythms. According to Elliott, everything moveswiththe same pattern as the tides-low tide followshigh tide, reaction fol-lowsaction. Timedoes not affect this scheme,because the structure ofthe market in its entirety remains constant.

In this chapter wewill reviewand analyze the followingconceptsofElliott: Nature's Law The "Secret ofthe Universe" The WavePrinciple Interpretative market letters.The focus of this section will be on the main areas ofElliott's

work that have long-lasting value. Even ifwedo not agree with some 11


16/94

12 THE ELLIOTT WAVE PRINCIPLE IN A NUTSHELL THE ELLIOTT WAVE PRINCIPLE IN A NUTSHELL 13

C D @ ajor waves

3 2 4 4 CAB 3 2G) 2 BA 5 44 CB 0 33 2 5A

4 C 2 Intermediate waves

5 3 2 4G) 25 24 C

3 0 5 42 C 32 5

Minor waves

of Elliott's findings, he must be admired for his ideas. We know howdifficult it is to create a new concept without the technical support wehave available today.When webegan to study Elliott's workin 1977, itwas a tremendous struggle to get the data needed for an indepth anal-ysis. Howmuch more difficult must it have been for Elliott when hestarted his work!With the computer technology available today, wehave the ability to test and analyze quickly, but it is still necessary tohave Elliott's ideas in order to begin.Elliott wrote, "Nature's Law embraces the most important of allelements, timing. Nature's Law is not a system, or method ofplayingthe market, but it is a phenomenonwhich appears tomark the progressofall human activities. Its application to forecasting is revolutionary.~*

Elliott based his discoveries on Nature's Law. He notes, "Thislaw behind the market can only be discovered when the market isviewed in its proper light and then is analyzed from this approach.Simply put, the stock market is a creation of man and therefore re-flects human idiosyncrasy" (Elliott, p. 40).

This chance to forecast price movesmotivates legions ofanalyststo workdayand night. Wewill focus onthe ability to forecast, and tryto answer whether or not it is possible.

Elliott was very specific when he introduced his concept. Hesaid, "All human activities have three distinctive features, pattern,time and ratio, all ofwhich observe the Fibonacci summation series"(Elliott, p. 48).

Once the wavescan be interpreted, the knowledgemay be appliedto any movement as the same rules apply to the price ofstocks, bonds,grains, cotton, and coffee. The most important of the three factors ispattern. Apattern is alwaysin progress, formingoverand over.Usually,but not invariably, you can visualize in advance the type of pattern.Elliott describes this market cycle as, "... divided primarily into'Bull Market' and 'Bear Market." (Elliott, p. 48). Figure 2-1 subdi-vides the bull market into five"major waves"and the bear market intothree major waves.The major waves 1,3, and 5 of the bull market assubdivided into five "intermediate waves" each. Then each ofthe in-termediate waves 1,3, and 5 subdivides into five "minor waves."

Figure 2-1 Elliott subdivided the "perfect" stock market cycle into major, inter-mediate, and minor waves.

*The Complete Writing" "fR.N. E ll io tt with Practica l Appl icat ion from J.R. Hill. J.R.Hill. Commodity Research Institute. N.Carolina. 1979 (subsequent references will citeElliott), p. 84.

The problemwith this general market concept is that, most ofthetime, there are no regular 5-wave swings. More often, the 5-waveswing is the exception. Tofine tune the concept, Elliott introduced aseries ofmarket patterns which take care of almost every situation.The most important ones are described next.


17/94

14 THE ELLIOTT WAVE PRINCIPLE IN A NUTSHELL

ELLIOTT'S MARKET PATTERNS

The 5-Waye SwingIfthe market rhythm is regular, wave 2 will not retrace to the be-ginning of wave 1, and wave 4 willnot correct lower than the top ofwave 1. If' it does, the wave count must be adjusted (Figure 2-2).- l : J - - .

2(a) (b)

6--*----A--//y ;; ~ nc)Figure 2-2 (a) Er roneous count ing in a 5:wa~e up- swing ; (b) cor~ec~ counting ina 3-wave up-swing; (c) erroneous counting In a 5-wave up-swing: (d) correctcounting in a 5-wave up-swing,

CorrectionsEach of the corrective waves 2 and 4, can be subdivided into threewavesofa smaller degree. The correction wave2 and corrective wave4alternate in pattern. Elliott called this the rule of alternation. Thismeans that, ifwave2 is simple, wave4will be complex, and vice versa.This is shown in Figure 2-3.

)3

4 42 2

la) (b)Figure 2-34 is simple.

(a) Wave 2 is simple; wave 4 is complex; (b) wave 2 is complex; wave

ELLIOTT'S MARKET PATTERNS 15

With this remarkable observation, Elliott linked Nature's Lawwith human behavior. In the sunflower, pinecone, or pineapple thereare spirals that alternate by first turning clockwise and then counter-clockwise. This alternation repeats itself in the corrective wave 2 andwave 4.

There are three types ofcorrections:1. Zjgzag in a downtrend (the opposite for the uptrend).

b

a A

(a) Minor c ccb) Intermediate ( el M a io r

2. Flats in a downtrend (the opposite for the uptrend).

2

4 3A c 5C a) Minor (bl Intermediate (e) Maior3. Triangles in an uptrend or downtrend.

,I

\ ~_1 3 5_-, ,, ,, ,-, -...--4- ,_: .-----.-- - - 2 ~\~5

(a) Uptrend (b) Oo>Nnt rend

"The student cannot be certain that a triangle is forming untilthe fifth wavehas started," observed Elliott (p,53).This makes it verydifficult to forecast price moves.

f


18/94


Elliott noticed that the standard type of corrections were notenough to cover all market actions. Therefore he added the complexcorrections, which he describes as:1. Minor correction ofthree waves

e

2. Double sideways correction with seven waves

3 7

3. Triple sideways corrections with seven waves.,,

10,,," 3 9 117

Elliott made the statement, "It is possible however to knowwhenelongated wavec will occur by understanding the rule ofalternation"(p. 51). It is not clear that it is possible to forecast wave c; and if itwere, it is not obvious that this rule is of any value for the investor.The corrections are so complex that it is impossible to determine, inadvance, any ofthe following essential points: The length of wave c The current state ofthe correction (i.e., isthis a simple, doubleortriple formation) What wave will come next.Elliott neverindicated either a definitive entry or exit rule to use

when trading. This means that the trader must use subjectivity andinitiative to implement Elliott's ideas.

,:~ !.

.


19/94

1B THE ELLIOTT WAVE PRINCIPLE IN A NUTSHELL FIBONACCI RATIO 199 \ 2 1,1,2,3,5,8,13,21,34,55,89,1447

8 4 Dividing any number of the Fibonacci sequence by the nexthigher number in the series causes the series to convergeasymptoti-cally to the ratio:

56 3 63 5 8

4 1.618 (1: 1.618 = 0.618)72

(a) (b) 9 Elliott recognized the importance for this summation series andhe wrote, "From experience I have learned that 144 is the highestnumber ofpractical value. In a complete cycleofthe stockmarket, thenumber ofminor waves is 144."The breakdown ofthis complete cycleis shown in the Table 2-1.

The entire Fibonacci summation series is employedhere. Elliottstated, "The length ofwavesmay vary, but not the number ofwaves.The numbers ofthis series are useful in timing the waves, both ad-vancing and declining" (pp. 45, 129).

The general application ofthis principle is that a movein a par-ticular direction should continue until such time that it completes anumber consistent with the Fibonacci summation series. This phe-nomenon is best illustrated in Figure 2-7.

Amovethat extends itself beyond 3 days should not reverse until5 days are reached. Amovethat exceeds 5 days should last 8 days. Atrend of9 days should not finish before 13days, and $0on. This basicstructure of calculating trend changes applies equally for hourly,daily, weekly and monthly data. But this is only an "ideal model,"andone must never expect commodity prices to behave in such a preciseand predictable manner. Elliott noted in his Nature's Law that devia-tions can occur in both time and amplitude, and individual waves arenot likely to always developin this regular pattern.

Figure 2-6 (a) 9-wave extension in uptrend; (b) 9-wave extension in downtrend.

They are doublyretraced, that is, a correction will pass overthesame ground twice, downand up.

Ifextensions occur in wave3, double retracement will be satis-fied bywave4 and wave 5.

All ofthe possibi lities that Elliott introduced and described inhis publications have not been shown here. The purpose of this re-view was to show the essence of Elliott's ideas and follow them asthey become more intricate. In its most complexstages, it is evendif-ficult for an experienced Elliott followerto apply these rules to real-time trading.

Elliott admitted, "Corrections in bull and bear swings are moredifficult to learn" (p. 48). The problemseems to be that the complexna-ture ofthe wavestructure does not leave roomto forecast future pricemovesin advance. It looks perfect in retrospect. The multitude ofrulesand situations described byElliott can beused to fit any price patternafter the fact. But that is not enough.

FIBONACCI RATIOTable 2-1Elliott is given the most credit for introducing the Fibonacci summa-

tion series as an investment tool. Hewrote, "Later I found that the ba-sis ofmydiscoveries was a law ofNature known to the designers oftheGreat Pyramid 'Gizeh' which may have been constructed 5000 yearsago.Fibonacci visited Egypt and onhis return disclosed a SummationSeries" (Elliott, p. 42). Chapter 1 presented the series:

Bull BearNumber of Market MarketMajor waves 5 3Intermediate waves 21 13Minor waves 89 55

Total8 completewaves34 completewaves144 complete waves


20/94

20 THE ELLIOTT WAVE PRINCIPLE IN A NUTSHElL

I :+--------------------------8---------------------------

----------- 5----------------+--- 3 --____,.34------------+--------------.-5-+------------------- 21 ------------------ ok---- 13---_

3-~--- 5-- 3_E--- 5----0';""-"'---- 8---+--..........;*


21/94


Elliott said, "The Principle has been carefully tested and usedsuccessfully bysubscribers in forecasting market movements" (p, 107).In another place hementioned, "Hereafter letters will beissued oncom-pletion ofa waveand not await the entire cycle.In this matter, studentsmay learn how to dotheir own forecasting and at noexpense. The phe-nomenon and its practical application becomeincreasingly interestingbecause the market continually unfolds newexamples towhich may beapplied unchanging rules" (p, 137).

My own work with the Elliott concept, from many different an-gles over 15years, doesnot support the contention that the wavestruc-ture has forecasting ability. The wave structure is too complex,especially in the corrective waves.The rule ofalternation is extremelyhelpful, but this does not tell us, for example, whether to expect: A correction of 3 waves. A double sideways correction. A triple sidewaysmovement.It is evenmore unlikely that any 5-wavepattern can be forecast.

The integration ofextensions in wave 1,wave3, orwave5 complicatesthe problem evenmore. The beauty ofworking with the Elliott conceptis not the wave count. Wecan only agree when J. R. Hill writes in hispractical applications, "The concept presented are extremely usefulbut have literally driven men 'up the wall' as they try to fit chart pat-terns to exactness in conformity with the Elliott wave"(Elliott, p. 33).

Elliott is focusedonpattern recognition. His wholeworkis stream-lined to forecast future price movesbased onexisting patterns. Hedoesnot appear to havesucceeded in this area. Elliott expressed uncertaintyabout his wavecount himself when he wrote in different newsletters,"The fiveweeks sidewaysmovementwas devoidofpattern [-1a featurenever before noted" (Elliott, p. 167). "The pattern of the movementacross the bottom is soexceedingly rare that nomention thereof appearsin the Treatise. The details baffle any count" (p. 165). "The time ele-ment [Fibonacci sequence] as an independent device,however,continuesto be baffling when attempts are made to apply any known rule ofse-quence to trend duration" (p. 180). "Thetime element isbased onthe Fi-bonacci Summation Series but has its limitations and can be used onlyas an adjunct ofThe WavePrinciple" (Elliott, p. 186).

Elliott did not realize that it is not the wave count that is impor-tant, but the Fibonacci ratio. Itis the Fibonacci ratio that represents

SUMMARY 23

c

A

B

Figure 2-10 Forecasting price move from point Bto point C not possible.

Nature's law and human behavior. This is what we try to measure inthe market swings. While the Fibonacci ratio is constant, the wavecount is confusing.

By studying Elliott's publications carefully, a rule with forecast-ing value can be identified as "Acyclical pattern or measurement ofmass psychology is 5 waves upward and 3 waves downward, total 8waves. These patterns have forecasting value-when 5 waves upwardhave been completed, 3 waves down will follow, and vice versa" (El-liott, p. 112).Wecould not agree more with this statement. Most likely, Elliottdid not realize that his strategy had taken a complete shift. Elliott triesto forecast a price movefrom point B to point C based on market pat-terns. Weconsider this impossible, and Elliott has never given a rulethat showedthat he was able to dothis mechanically (Figure 2-10).

Elliott's latest statement takes exactly the opposite strategy. In-stead offorecasting a price move from point B to point C, he waits tothe end of a 5-wave move,because 3 waves in the opposite directioncan be expected (Figure 2-11).

34

-- - - - - . - j Forecastingis possible

c

Figure 2-11 Forecasti ng price move after t he end of a 5-wave cycle possi b Ie.


22/94


We totally agree with Elliott's approach here and will comple-ment the idea with additional rules in later sections. Wewill also in-troduce other investment strategies closely related to the Fibonacciratio. Elliott never workedwith a geometric approach. Wewill intro-duce logarithmic spirals as an investment tool. We strongly believethat this is the solution to the problemof combining 'price and time.'This has never been donebefore.

The concentration will be on Fibonacci. The following chapterswill work with daily and weekly data. Research shows that intradaydata can beapplied where only daily data has been used thus far. Morehistorical testa are needed before definitive rules can be set.

I :!~'I

1 '1

1

' , ' 1i I

3WORKING WITH A5-WAVE PATTERN

Returning to Elliott's original work, one of his most important state-ments is, "Acyclical pattern or measurement ofmass psychologyis 5wavesupward and 3wavesdownward,total[ing18 waves.Thesepatternshave forecasting value--when 5 waves upward have been completed, 3waves downwill followand viceversa" (p. 112).This appears to be theonlytime that Elliott gavea definitive rule with forecasting value, andmany times the market behaves exactly as this pattern dictates.

In this chapter, wewill analyze this great observation ofElliott,combining it with additional rules to give definitive entry and exitpoints. Consistent with the Elliott concept, the end ofthe fifth waveisconsidered to be a safe point to invest. The rare case ofan "extensionin the 5th wave"will be explained later.The five waves can be seen in charts of any time period, whetherintraday, daily, weekly, or monthly. The greatest problem for the in-vestor is waiting for the end ofthe fifth wave.To identify the five-waveswing, wemust return to Elliott who said: Under normal circumstances, wave 5 appears similar to wave 1. Most ofthe time"wave3 is the longest wave. Wave4 should not touch the top ofwave1 in an uptrend.

2 5

26 WORKING WITH A 5WAVE PATTERN


23/94

1 0

55 ~ Q ~ ' I Q ' ~ ' I, ~ 'I I ' I" I I ' Zl lo ' i' IQ ' J II O ~ l' o' 5 H ~~ I IO , ,1 ' o 'i II O ~ 1 11 i I [ I ' l il i a , 1 ! I I D , ~ ' I O ' 3 1 1 0 ' ~ I I c ' b lc' 4 1I a ' ! l l d Q I I I ~ I I I ' i l l { ~ I ' c t l o ~ I I c ' ~ l lo ' ~ ! l r

0 1 I I I b Z O ~5 ~ q O ~ 01 !J 1 1 n z . 3 D o ~ D ~ 1 3 I 1 Z ~ U O Z 0 1 I I 1 5 Z O Z ~ Z q O f O q 1 3 I I :S WI S S F R A NC S S / 9 ~ . M 6 3 . 1 - 2 . 9 ( 4.6r.): 6 9. 2 I : 9 . 1 1 J A I l O r - . 6 1 8 D I R : + 1Figure 3-1 5-wave swing on weekly Swiss franc chart from 07-07-89 to 06-30-92. (Source: TradeStatlon, Omega Research, lnc.)

Following this path, the weeklychart ofthe Swiss franc was cho-sen to demonstrate a 5-waveswing (Figure 3-1).

In Figure 3-1, the 5-waveswing is comprised of: Wave1 from 79.80 to 70.12 Wave2 from 70.12 to 74.31 Wave3 from 74.31 to 61.95 Wave4 from61.95 to 70.48 Wave5 from 70.48 to 55.25

PRED ICT ING THE END OFW AVE 5 US ING A TREND CHANNEL

Elliott tried to forecast the end ofwave5 byworking with trend chan-nels. In an upward movement offive waves. a base line is drawn acrossthe ends (lowest prices) of wave 2 and wave 4. A parallel line is thenI'

PREDICTING THE END OF WAVE 5 USING A TREND CHANNEL 27

5

Figure 3-2 Upper line of the trend channel is penetrated.

drawn across the highest point ofwave3 (Figure 3-2). When the fivewaves are complete, there are a number of options for entering themarket. The followingcases showthe different possibilities for invest-ing at the time wave 5 ends.

Case 1"Usually wave 5 will end approximately at the parallel line, when anarithmetic scale is used" (Elliott, p. 60).

If a trade is entered when the upper line of the trend channel istouched, it runs the risk ofthe market going evenhigher. Elliott nevergave a solution for protecting an investment against the risk of awrong analysis.

Case 2"Tfwave 5 exceeds the parallel line considerably and the compositionofwave 5 indicates that it has not completed its pattern. then the en-tire movement, from the beginning ofwave 1, should be graphed on asemi-log scale. The end ofwave5 may reach, but not exceedthe paral-leI line" (Elliott, p. 60).

If the market price reaches the trend channel ofa semi-log chart(Figure 3-3a), it is positioned at a very goodpoint to invest, but onlywhen combined with the conservative entry and exit rules introducedlater. The onlyproblem isthat this situation is very rare and not likelyto be reached. Waiting unt.iJ the price touches the upper trend lineruns the risk ofmissing the moveentirely (Figure 3-3b).

PREDICTING THE END OF WAVE 5 WITH FIBONACCI RATIOS 29


24/94

28 WORKING WITH A 5-WAVE PATTERN

(a) (b)Figure 3-3 (a) Trendline on semi-log chart istouched; (b) trendline on semi-logchart isnot touched.

CaseJ"When 5 waves upward have been completed, 3 waves downwill fol-low."(Elliott, p. 112).Using this approach, wewould: Wait for 5 waves to be completed. Wait for the a,b correction. Invest when wavec penetrates the base line.

This strategy is shown in Figure 3-4. This is a very conservative ap-proach, and the pattern can be foundoverand overagain in the charts.The advantage ofthis strategy is that: The market has already changed trend direction at the time ofentry, and There is still good profit potential, measured by the combinedamplitude ofall fivewaves.(For example, Figure 3-5 showshowpowerful this signal can be in the Japanese yen.)

5b _....3

c

Figure 3-4 Complete cycle of 5 impulse waves and 3 corrective waves.

31001 0 t O

l l i O1110

., . r,,1100l b t O'S~Dl ~ b O13&0 , I L ! , ,11 ,LI! !III",! .!l l '1 1 1 ,1 !ILl :11' 'LI! 1111 ,!Ii , , ,1 !It! dll .111 I I! ! . ,, ' . 11 1 . 11 1 t il l , 'I I dill ,LI! !DI Dt 0 3 O ~ as Db a ltIa~ II I~ tb D~ II II tS 0 I DI 15 n 30 01 I~ tl t~ OS It I~ tb Db 11 t o tJ A P A N E S E Y E N 89/92.0 R A t i O : . 6 1 8 D I R = + 1Figure 3-5 Daily chart Japanese yen from 01-01-92 to 07-25-92. Sell signal atan a-b-c correction. (Source: TradeStation, Omega Research, Inc.)

The disadvantage ofthis strategy is that: Thetrading opportunity can bemissed ifthere isno a-b-c correc-tion,and It is valid only for major moves. Smaller swings can often havecorrections that are too small to be profitable.

PREDICTING THE END OFWAVE 5 WITH FIBONACCI RATIOS

The end ofwave5 may also be identified by integrating the Fibonacciratios into the analysis. Wehave already learned that there are tworatios in the Fibonacci Summation Series, 1.618 and 0.618, that canbe used.By integrating the Fibonacci ratios into the 5-wave pattern,shown in the weekly chart ofthe Swiss franc (see Figure 3-8), pricegoals can be calculated. But evenbefore that can be done, the swingsize must be defined.

30 WORKING WITH A 5-WAVE PATTERN PREDICTING THE END OF WAVE 5 WITH FIBONACCI RATIOS 31


25/94

Swing SizeAprice swing is continuous movement in one direction. Because pricesoften move up and down in very small increments, it is necessary toignore someof this "noise" by eliminating all movement that does notcontinue in one direction for at least a minimum point value. If, onsuccessive days, the Swiss franc were to move +50, +100, -30, -la,+70, -20, +25, and the minimum swing size was 50points, we wouldonly see an upwards moveof 185. The 40 and 20 point reversals wereless than the 50-point filter; therefore, they are ignored.

A minimum swing size is needed to make the best use of the Fi-bonacci ratios. For that reason, intraday charts and examples are notused in this book. Shorter time intervals contain more "noise" that is,price movement that is random and therefore not predictable. In addi-tion, the magnitude of that noise is relatively large compared to theswing size possible during a short interval. The inclusion ofmore noisewould obscure the useful information and make analysis ineffective.

Elliott noted, "In fast markets, the daily range is essential, andthe hourly useful, if not always essential. On the contrary, when thedaily range becomes obscure, due to slowspeed and long duration ofwaves, condensation into weekly range clarifies" (p. 139).

The intent ofthis book is not to present empirical tests for everycommodity, but to present a concept with adequate convincing data.As a guideline for examples, the minimum swing size for a daily chartof the Swiss franc, Deutsche mark, or Japanese yen will be 100 basispoints. Forthe British pound, it will be 200 basis points. When usingweekly charts, the swing size will be twice the size of that used onthe daily chart.

I I'! II;l Confirmation of the Swing Highs and Lows

A swing high or lowis confirmed when there is a counterswing of atleast the minimum swing size in the opposite direction. For example,the minimum swing size is 100 basis points, prices must decline byatleast 100 points (without being interrupted bya net upwards moveof100 points) to confirm a swing high. This is shown in Figure 3-6.Minimum Amplitude of the Total Five SwingsElliott noted that after a 5-wave swing pattern there will be a 3-swingcorrection. But this approach onlymakes sense if the amplitude ofthe5-swing pattern is reasonably big. Even though Elliott noted, "Timing

- r - - - - - - - ~100 IPoints tL _ _ _ _ _ t + C l o se b e l o w the lo wollhe highest da y

Figure 3-6 Confirmation of a swing high.is the most essential element. What to buy is important, but when tobuy is more important," (p,84) he never gave a clear rule for enteringand even more important, for liquidating a position.

Consider a perfect 5-wave count within a 200 basis point up-swing. It would be unrealistic to expect a profit from a short positionfollowing the strategy described in case 3.

The reason for this is as follows:The minimum correction allow-able using the Elliott concept is 38%(the transformed Fibonacci ratio0.618) ofthe total amplitude ofa 5-wave cycle.

This strategy leaves only a small profit margin for a move of200 basis points. Yet it is a perfect strategy-one of the best in allofElliott's work. If we have a larger amplitude of 1,000 basis points,

38%Ratraoemenl_ _ _ _ J _1000Points1Bf--~- - ----- -- --- t38% Ret racement200 Points - ---,-J__

A A (b)a)Figure 3-7 (a) 200-point swing with 38% retracement: (b) 1000-point swingwith 38% retracement.

32 WORKING WITH A 5-WAVE PATTERN WAVES 1, 2, AND 3, AND THE FIBONACCI RATIO 0.618 33


26/94

waiting for the end ofwave5 and selling into the a-b-c correction is averyinteresting strategy (Figure 3-7). This pattern can be seen in themarkets over and overagain.Wave 1 and the Fibonacci Ratio 1_618Elliott never gavedefinitive rules to apply his concept to the markets.Wewill try to introduce rules that make the Elliott concept safer totrade. The weeklySwiss franc (Figure 3-8) will be used as an exam-ple ofhowthe Fibonacci ratio can be used with wave 1.

Whenever there is a 3-swing pattern, the extreme point ofwave5can be calculated using the Fibonacci ratio 1.618 (Figure 3-9). Itisnot possible to knowwhether this extreme point will everbe reached,but we doknow that this precalculated price level is very importantshould the prices reach this level.The followingshowshowto calculatethe end of wave 5 using the ratio 1.618 applied to the weekly Swissfranc chart (Figure 3-8);

Start of wave 1 79.20 9.08 X 1.618 = 14.69Bottom of wave 1 70.12 70.12 - 14.69 = 55.43 priceDifference 9.08 points target

Then, the real time bottom ofwave5 was 55.25.

Figure 3-8 Weekly chart Swiss franc from 07-07-89 to 06-30-92. The ratio1.618 isused with wave 1 to calculate the end ofwave 5. (Source: TradeStation,Omega Research, Inc.)

- r - - - - - - - - - - - - - - - - -1.618 @

C D @1.0__--L

@ 11.618L _@a (a) (b)

Figure3-9 (a) The ratio 1.618 isused with wave 1 to calculate the end ofwave5; (b) the ratio 1.618 is used with wave a to calculate the end of wave c in acorrection.

THE AMPLITUDE OF WAVES 1, 2, AND 3,AND THE FIBONACCI RATIO 0.618Whenever the peak of wave3 is established, the end ofwave 5 can beprecalculated using the ratio 0.618 (Figure 3-10). Howdo we knowthat the peak ofwave3 is established? Wave3 must to be longer than wave 1,and In an uptrend, wave 4 should not go lower than the bottom ofwave2 (the opposite for a downtrend) (Figure 3-10).

:;: @- t - - - - - - - - - - -0.618@ r1.0 @- - - - - - - ~ - - - _ ! -

figure 3-10 The total amplitude of the wave 1r 2,3, isusedwith the ratio 0.618to calculate the end of wave 5.

34 WORKING WITH A 5-WAVE PATTERN WAVES 1,2, AND 3, AND THE FIBONACCI RATIO 0.618 35


27/94

By analyzing the weekly Swiss franc chart, the end ofwave5 canbe calculated based on the total amplitude of the first three wavesmultiplied by0.618 as follows:

Start ofwave 1Bottom ofwave 3Difference17.26 X 0.618 =61.96 - 10.66 =

79.2061.9517.26 points10.6651.29 price target

Two price targets can then calculated for end ofwave5, using the tworatios 1.618 and 0.618:

Amplitude ofwave 1 X 1.618 =Amplitude ofall 3 waves X 0.618 =

55.25 (see p. 32)51.29

Do we know whether the forecasted price level will ever bereached? Absolutely not. Wecan never knowin advance whether theselevelswill be reached. But should these levels everbe reached, there isan excellent chance ofa trend change.It is better when the precalculated price targets are close.But in reality this happens very rarely. Most of the time there is aprice target band-the difference between the two targets (55.25 and51.29)-as seen in Figure 3-8. Whenever this price band exists, thetrader must decide where and when to enter the market.

The entry rule must be integrated into the analysis. To completethe strategy, a stop-loss rule must also be introduced, as wellas a re-entry rule, profit target and trailing stops. Each of these steps will bedescribed in detail.

Entry Rule. Whenever a 5-wave swing has been identified, a confir-mation of the trend change must occur before entering the market.The following two situations are possible:1. After waiting for the precalculated price targets to be reached,the market reverses a little too early and the trade ismissed. Wedonot want "chase" the price.

2. After waiting patiently, the buy order gets filled in the fallingmarket. Instead of reversing exactly at the precalculated price,the market keeps falling.

11

. : ~ :J

1:'

These dilemmas can never be totally solved, but our experiencetells us that the effect can be smoothed byworking with an entry rule.The trade-off in working with an entry rule is that part of the movewhichwewant to capture must be sacrificed. But it has the advantageof not entering the market too early and not suffering losses that re-sult from an extension in wave 5. (This case will also be discussed inmore detail later.)

The entry rule is based on the findings of Elliott, which indicatethat, after every 5-wavepattern, there is an a-b-c correction, or a doubleretracement. Once wavea and wave b are completed, we can sell intowave c if the previous valley is broken (Figure 3-11). The opposite istrue for a buy signal.)

0t 5 + + t + t @0 C D t t @3 . ' t Iella4 c @ t t f t - - + - f B " Yt -I -

2 + @(a) (o l

figure 3-11 (a) After the end of a 5-wave pattern, we can sell into wave c:(b) after the end of a 5-wave pattern, we can buy into wave c.

This is a very conservative approach. In our opinion, it is one ofthe most important ofElliott's discoveries. This pattern can be foundon the daily chart of the Japanese yen (Figure 3-12) as it develops.

The approach is conservative because it has the disadvantagethat the trend change might be completelymissed if there is no a-b-ccorrection, but a strong trend reversal instead. More aggressive in-vestors whowant to buy may dobetter entering an order in advance ofthe point where the close is higher than the high of the lowday (andthe opposite for a sell signal), as shown in Figure 3-13.

Again, this is a more aggressive approach. It runs the risk ofget-ting stopped out when there is a wave 5 extension. A trader must be

36 WORKING WITH A 5WAVE PATTERN WAVES1,2, AND 3, AND THE fiBONACCI RATIO 0.618 37


28/94

1100I O Z O

11bO11101100

15~D

1310,i/U (1 / ,,0 c/ O dO f il l ' 11 ' o I I! d ll "II i/ O dt) lIi l o J d,J uO ,O ! .HI , I I ' rill! .. l I'll II II II IJ lo I O~ 03 O ~ 05 Db D1nD~ II I ~ % b O ~ II 1!!5 01 01 15 n 3D 01 I~ Z I % ~ 05 IZ Iq 2D Db I J ZC ZJ A P A N E S E Y E N 9 9 / 9 2 . D R A t i O : . 6 1 8 D I R : t l

figure 3-12 Sell signal after an a-b-c correction. (Source: TradeStation, OmegaResearch. lnc.)prepared for a string of losing trades when a double or triple correc-tion follows the end of the 5th wave. The selection of entry rules de-pends on the risk preference ofthe investor.Stop-Loss Rule. Whenever a position is entered, it should beprotectedwith a stop loss. One wayofplacing a stop lossis to use e.price square.This technique can be applied to all commodities, all charts, and iseasy to use.

ii,1 ' 1 . :' I : ' "IIII: II''! t I + ~ I Closehighert T than high of1 t + ~ t ffg> " : - J day3 I lowpay_ j J _ - - t -+ + +Lowestlow 5(a)

Low of~~igjdayt . l . t + -+ t + t lt+ Closelower1 t t tt than l ow oft t + + 2 highestday

(b)figure 3-13 (a) Buy signal when Ihe close ish igher than the high of the lowestday; (b) sell signal when the dose is lower than the low of the highest day.

: j} One pricel=~4=R~ t=*-=- ' : p squarestop loss-ne week inl imeFigure 3-14 Stop loss a "square" above/below the entry point.

.".

To create a price square, measure the distance of five businessdays (for a daily chart) or five weeks (on a weekly chart). Apply thisdistance "in time" to the price scale, that is, take the horizontal timemeasurement and use the same distance vertically for a price stoploss. This gives the stop loss in points on every chart (Figure 3-14).Because the same measurement is used for both horizontal and verti-cal, the result is considered a "square." Place the stop loss (the pricesquare): Above the high of the highest day prior to the short signal, or Belowthe lowofthe lowest prior to the long signal.

... -.

The market must close above/belowthe square to be stopped out.There are times when the result ofthis stop-loss rule is a price that istoo far from the entry point for a trader, whowould rather get stoppedout more frequently with small losses than take fewer large losses.One solution is to place the stop loss one tick above/belowthe previouspeak/valley after entering the market (Figure 3-15).

Neither of the stop-loss selections is better or worse than theother. It is the risk preference of the individual investor which deter-mines which stop is better one. The most important strategy is to beconsistent. Once a strategy is chosen by some tested method, stickwith it.

' . ' ; :~ 5 b S t op- loss leve l- - se J 13 a

4 c2

figure 3-15 Slop loss above the previous peak.

if!,"i'i!


29/94

Ii';

i,


Buy signal (Close higher/ than high of l owest day)+ t t t+ 1Stop L o S S / + _ Re-entry (Close higherthan high of lowest day)+

Figure 3-16 Re-enter after s top Joss when close isagain higher than high of low-est day.

Re-Entry Rule. Whenever a posit ion is stopped out, it is re-entered us-ing the following rule (shown in Figure 3-16). This re-entry rule isthe same as already described for the entry rule. After a short posi-tion is stopped out, it may be re-entered on the short side when theclose is lower than the low of the highest day (the opposite for a longposition).P ro fi t- T ar ge t R u Je . When either a long or short position is held, it canbe liquidated at a profit target whenever that precalculated value isreached. Using the Swiss franc weekly chart (Figure 3-8) as an exam-ple, the profit target is calculated as follows:1. On Figure 3-8, the weekly Swiss franc chart, the 5-wave swingstarts at 79.80, with the low of the 5-wave swing at 55.25.

2. Using a profit target of38% (the complement of the Fibonacci ra-tio .618), the calculation is as follows:

79.80 - 55.25 = 24.5538% of 24.55 = 9.3255.25 + 9.32 = 64.57, the target price

For a long position, the profit target is the 38% retracement ofthe distance from the high to the low. For a short position, the profittarget would be a 38% retracement from the low to the high. This caseis shown in Figure 3-17.

WAVES 1, 2, AND 3, AND THE FIBONACCI RATIO 0.618 39

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - High

2

438%, - - - _ F irst profit

Correction targeta ~ " " , L5

figure 3-17 The profi t target is 38% the dis tance between high and low.

3

A correction of 38% is the minimum that can be expected after acompleted 5-wave swing. Most of the time, a 50% correction is morelikely.Trailing Stops. The rule most important to successful commoditytrading is "Never give back all of your profits." How many times has ithappened that a comfortable profit in a position is lost in a single"blowout," and ends up in a loss? To protect these profits, and as analternative or addition to the profit target, a trailing stop can be used.This strategy has the advantages that: Most of the profits that develop can be protected, and We can still participate in a trending market.Of course, the position might get stopped out before the profit

target is reached. The decision to use profit targets, the defensivestrategy of a trailing stop, or a combination of the two, is an issue ofthe individual investor's risk preference.

A trail ing stop of four days (Figure 3-18), protects the positionand offers a good chance of capturing a market trend.

An alternative to a trailing stop (in this example, a long position isheld) is to place the stop loss below the previous valley (Figure 3-19).Whenever there is a new, higher valley, the stop is raised. In a trendingmarket, the position might be held for a long time.

There is no "best" stop-loss rule. Both of these have their advan-tages and disadvantages. It is most important that stops are used toprotect profits and reduce risk.

r


30/94


f I 21 + t t 4t + Bu y signal + ttlt + ~ + -I - t + L - .j :- - 4-day trailing stop3 f + t ~ _ + _ t + tf \ Close higher than5 high of lowest da y

Figure 3-18 Four-day trailing stop offers a good chance ofcapturing part ofa mar-ket trend.

figure 3-19 Trailingstop on previous valier.

INVESTING WITH OPTIONSI.: IOptions can easily be used as an alternative strategy to futures, butonly if the "right" strike price and expiration month are chosen. Oneproblemwith options is that the options premium alwaysworks againstthe buyer. As the selected expiration is farther out, the premium in-creases due to the "time value."

Without goinginto toomuchdetail, the followingideasmight im-prove trading performance byusing options instead offutures: When workingwithin a price band, onlybuy an optionwhen thesecond price band is reached (Figure 3-20).

;I,;'} i

INVESTING WITH OPTIONS 41

B u y put options-, -- --- --- -- --- --- -- --- -- 2ndprioeband0.618 5- r - - - - - - - - - ~ - t - - - - - - - - - - - - - - - - - - - - - - - 1st price banCI

1.618

.... :; .. 2 J. '~,

4

Figure3-20 Buyof put options when the 2nd price band is reached.

Ifthe secondprice band is not reached, wait for wavea and waveb to be completed, then invest in wavec followingthe entry rule.That is, buy or sell when the previous four highsllows are brokenin wavec (Figure 3-21).The advantage of working with precalculated price targets is

that wecan buy a call into a falling market or buy a put into a risingmarket. This reduces the premium of the call or the put substan-tially. An entry rule is not needed to reduce the risk of the position.Figure 3-22 shows the application of an entry signal with a call op-tion futures position.

~'-

Buy put options4

Figure 3-21 Buy of put options when the 1st price band is reached.

SUMMARY 43


31/94

42 WORKING WITH A 5WAVE PATTERN

(b) 3

- - - - +) 0 t tJ . t + t1 + + t Buy M " , , ,1.618 + t + + \tI f - ~ I - - Tf + t--- -- - -- --- ~- - -~------

Figure 3-22 (a) The ratio 1.618 isused to calculate the end ofwave 3. Buycall op-tions when the target price ispenetrated; (b) the ratio 1.618 isused to calculate theend of wave 3. Buyfutures when the dose is higher than the high ofthe lowest day.

Buying a call option eliminates the risk ofgetting stopped out.Tobe sure that the optionmovesin tandem with the futures price, al-ways buy the call with a strike price at the moneyor in the money.Theexpiration should be about three months away from the time ofentry.Ifthe expiration month is too far away, the optionmight not run dollarfor dollar with the futures price. Ifthe expiration month is too close,the option might expire before the trade is completed.

Options should be substituted for futures only after big priceswings. In the Swiss franc, Deutsche mark, or Japanese yen thisshould be about 10 full points (50.00 to 60.00) measured from thehighest high to the lowest low in the 5-waveswing. A big retracementcan only be expected in conjunction with a larger.

SUMMARYThe Elliott concept is most famous for the definition of the 5-wavepattern. In a "regular" market, itworks with stunning accuracy. butthis is not reality. Mostof the time there are only "irregular" marketpatterns.

Weconsider it impossible to predict orforecast the beginning orend of a 5-wave pattern. But there are very good chances to profit by

i l l :1 '1 ;:1" I'

waiting until a 5-wave pattern has developed. Weagree with Elliottwhe~ he said, "Acyclical pattern or measurement ofmass psychol-ogy IS 5 waves upward and 3 waves downward, total 8 waves. Thesepatterns have forecasting value; when 5 waves upward have beencompleted, 3 waves downwill follow, and vice versa" (p. 112). But theproblem and risk in this strategy is ofan extension in the fifth waveif there was no extension in wave lor wave 3 (extensions will be die-cussed later).Elliott acknowledged someproblemsin his conceptwhen hewrote"Exhaustive research ofall available records doesnot discloseany simi-lar abnormalities tothose of the past six months, such as: Omission ofan intermediate wave(wave5ofthe late bear market), Two extensions in one intermediate or Distortion ofpatterns ofthe averages" (p. 171).Elliott must be admired for the tremendous effort to implement

his concept with the limited historical data and tools available at thattime. But wecan not agree with his vision to predict future price movesin general. There are somespecial rules that can be filtered out. Theserules bythemselves will give Elliott a pJacewith the great analysts ofour century.

The combination ofweekly and daily data can improve the qual-ity ofthe analysis. In order to reduce risk, guidelines were introduced.inthe situation where wave5 ends at a point.corresponding to the cal-culation of the Fibonacci ratios 1.618 and 0.618. These guidelinesincluded entry rules to determine entry timing. But entering the mar-ket is only part ofthe investment strategy. Wealso introduced: Stop-loss rules to limit risk. Profit targets. Trailing stops to capture trend profits.Wehave seen that. a 5-wave pattern with small amplitude has a

small profit potential in the correction. The bigger 5-wave swingshave substantial profit opportunity, but occur rarely.

The risk preference ofthe investor is the most important factor inselecting trading rules.

'1.. i 44 WORKING WITH A 5-WAVE PATTERN


32/94

. I Options on futures are a very good alternative to outright fu-tures positions when price swings are big; their use takes most ofthepressure off finding the perfect entry point. Using an option, we canbuy or sell against the market trend, keep the premiums lowand havea limited risk. The option should have a strike price at or in the money,and an expiration date of at least three full months.

i:

'Ii

4WORKING WITHCORRECTIONS

According to Elliott, every market moves in rhythms. Impulse waves,defining the major price movement, will have a corrective wavebeforethe next impulse wavereaches newterritory. Elliott stated, "Correctionsin both bull and bear markets swings are moredifficult to learn" (p,48).Beforebeginning the analysis ofcorrections, let us briefly reviewwhathas been coveredin detail in the previous chapters.

Elliott identified three types ofcorrections:1. Zigzags,2. Flats, and3. Triangles.In addition to these, there are the complexcorrections:

4. Double sidewayscorrections with 7 sidewayswaves, and5. Triple sidewayscorrections with 11 sidewayswaves.Even though Elliott made the statement, "It is possible however

to know when an elongated wave c will occur by understanding the

45

46 WORKING WITH CORRECTIONS RULESTHAT ARE RELIABLE 47


33/94

rule of alternation" (p. 51), it-does not seem likely that the analysis ofcorrections have forecasting value. Itmight be possible to see wave cdevelop based on the rule of alternation, but this fails to tell us: Whether we are in a single, double, or triple correction, How far wave c will go, or When to enter and exit the position.On the bottom line, the Elliott concept must make money; it is

not just a pleasant mental game. Unfortunately, Elliott does not givethe definitive rules for corrections that would allow a trader to reachthis goal. The complexity of the corrections leave too much room forsubjective decisions.

Used here, forecasting does not mean a small breakout of a chartpattern, but the ability to predict a significant price move or change ofdirection.

Elliott appears to have gotten caught in the complexity of his ownconcept. The more he noticed the exceptions and tried to integrate theminto his concept, the further he moved away from definitive rules.

RULES THAT ARE RELIABLEThere are two definitive rules in the Elliott concept that are reliable:1.Wave 3 is normally the longest wave.2. In any impulse or corrective market cycle, we have at least 3waves.

But these rules get buried in the wave patterns. They are not aschallenging as forecasting a price move from point A to point B, butthey seem to be extremely reliable and can be applied to any kind ofmarket and commodity.

Following the idealized Elliott concept, the breakdown of a pri-mary market cycle is shown in Figure 4-1, and expressed as: Waves 1, 3, 5, a, and c can be subdivided into 5 waves of lesserdegree. Waves 2, 4, and b can be subdivided into 3 waves oflesser degree.

5 cIV

Figure 4-1 Breakdown of a primary market cycle in the idealized Elliottconcept.

Elliott's two great discoveries were expressed when he said, "Aswill have been noted, all corrective movements regardless of degreeare composed of 3 waves," and "It is important to note that wave 3 isnever shorter than both wave 1 and wave 5." (pp, 56,57). From thesestatements we can draw the conclusion that a future price move can beforecasted if:

We only concentrate on a 3-wave swing, Invest only into the correction of wave 2, and Take profits on precalculated price targets.

In this sense, "forecasting" means the ability to identify, in ad-vance, a price move that should allow profits to be taken at a precalcu-lated profit target following specific rules (to be provided later in thischapter).

Figure 4-2 shows profitable areas ofirrvestment marked PI, P2,P3, P4, P5, P6, P7, and P8.

I,

48 WORKING WITH CORRECTIONS WHEN NOT TO INVEST 49


34/94

Figure 4-2 Profitable areas for investment based on the Ell io tt concept .

A 3-wave swing can be expected, most often when the marketchanges trend direction. There should be goodprofit potential if: Wave1has achieved the minimum swing size. Wave3 is longer than wave 1. Weinvest in the correction ofwave2. Weworkwith a profit target. Wealways protect our position with a stop loss set at the begin-ning ofwave1.This strategy seeks only safe profits froma very special subset of

all themarket patterns available. It isdrawn fromthe part ofthe Elliottconcept whichoffers forecasting possibilities within its original rules.But our choiceof forecasting is completely different from Elliott's. Wehaveconcentrated ona very special combinationofevents,whileElliottclaims to forecast all 5-wavemoves.There is noindication that Elliott'smethod can be successful.

It should be noted that the minimum swing size is a constraintonthis strategy, but a necessary one.Without a minimum swing size,

the profit potential of many trades will be too small and the resultwouldbe frequent whipsaws.

There is also a maximum swing size. If there is an initial wave 1(for example, 2,000 basis points on the British pound weekly chart),it is highly unlikely that either a correction or wave 3 will developwithin a reasonable period oftime. Elliott said, "On a weekly rangechart, all wavesofthe cycle are clear except those ofwave#1, the sub-divisions of which are clear in the daily chart. In fast markets thedaily range is essentiaL On the contrary when the daily range be-comesobscure, due to slowspeed and longduration ofwaves, conden-sations into weekly clarifies" (p. 139).

There is no indication that all waves will be clear on a weeklychart. Experience doesnot confirm this. The problemwith big swingsin weekly charts, especially in the currencies, is that there are upperand lower bands in which the currencies normally move relative toeach other. The Deutsche mark will not move50 pfennige against thedollar, even if a 5-wavecount predicts it.

WHEN NOT TO INVEST

Following the Elliott concept, wewould not buy in an uptrend at theend of wave 3. In Figure 4-3a this point is marked A, and in Fig-ure 4-3b it is marked B. The reason isthat the Elliott concept cannotresolve the dilemma ofwhether:

A correction is onlypart ofa long-term trend, or A correction is the beginning of a new trend in the oppositedirection.

3 " 3 BA4 4

2 2(a) (b)Figure 4-3 (a) No buy at the end of wave 3; (b) no buy at point B.

50 WORKING WITH CORRECTIONS SIZE OF CORRECTIONS 51


35/94

u'"21; O n 23 2" J 2

X 5 5 2

21

Figure 4-4 Areas of uncertainty out of the Elliott concept.

Note that the most critical point is at the end ofwave3. Will thefinal wave5 developat point A in Figure 4-3a, or isit the beginning ofa bigger correction wave as shown in point B ofFigure 4-3b? Elliottfollowersmight argue that a more detailed wavecount or the integra-tion of weekly, daily, and intraday charts might help here. It is notlikely, for the complexity of the wave count leaves too much room forerror.

In the idealized chart ofthe Elliott concept shown in Figure 4-4,areas were marked as X], X2, X3, X 4,XS. Arguments in favorofElliott'stheory are based onthis idealized chart shownin Figure 4-4, but in re-ality, these chart patterns very rarely occur. Reality is Figure 4-5, aplot ofthe DowJones Industrials between 1960 and 1980. This repre-sents a typical sidewaysmarket and can be seen in its many variationsonintraday, daily, and weeklyversions.Applying the DowJones patternto the mechanical rules ofthe Elliott concept seems to be impossible.

SIZE OF CORRECTIONSThe most common approach to working with corrections relates thesize ofthe correction to a percentage ofthe prior move.This approach

TRENDS IN COMMODITY & SECURITIES PRICESao:r-:---".=


36/94

3----- --- -38%- ----- ---5()Ok

______________ 2 4 l ~ %Figure4-6 Sizeof corrections 38%, 50%,62%.

or even 100%of the original price moveat the beginning ofwave1. In-vesting tooearly increases the chance ofloss bybeing stopped out, eventhough the market may ultimately movein the direction originally ex-pected. Waiting for a specific retracement, for example 62'/(}mightcause the wholemoveto be missed.In general, the safest correction in which to invest depends on: The investment strategy (e.g., buyandholdor short-term trading). The volatility ofthe commodity. The size ofthe swing upon which the retracement is measured. The strength ofthe trend. The type ofdata used (i.e., monthly, weekly,daily, or intraday).

:3 Profit-~----------- --target0.618 lawIProfitpotential10 -- 38%. _ L_ _ _ _ _ _ J _ 2 Correction

(a)

3 Prom-~- ----- --- -- -- target0.618 ifMIl

Profitpotential__ 62% J _2 Correction(bl

Figure4-7 (a) Profit potential after 38% correction; (b) profit potential after62% correction.

CORRECTIONS IN A LONG-TERM TRENDItis generally accepted that trading the correction to a price movecanreduce risk. This strategy became very popular in the 1980s for buy-ing securities. By betting on a long-term uptrend, it was easy to buyinto the corrections ofthe DowJones or S&P 500. They could alwaysbe expected to make new highs, as seen in Figure 4-8. There aremany success stories.Problems with this strategy start when there is nolong-term up-trend. Looking at the DowJones over the longer period from 1970 to1980, the pattern of price movement is much more sideways. Therewere a large number ofbig swings ranging fromabout 500 and 1,000points (Figure 4-9). It is inevitable that patterns such as these willcontinually reappear.

[t.] 1"IDt-j,11-L'I '* "S 5" 4 0(1 00

IWVr.l'0- 32"11~ ,H 3280SL'~(IE 31141 , ' S " S Z 5"e3 1"" 1~ 11'66 1:=87 t'sg U'~ .l1'i"O t"" l IWZFigure 4-8 Monthly Dow Jones 30 Industrials Chart from 1982 to 1992.(Source: Commodity Quote Graphics, 1992.)

54 WORKING WITH CORRECTIONS CORRECTIONS IN A SHORTTERM TREND 55


37/94

ii,1 1

--~t ~ TRENDS IN COMMODJTY & SECURITIES PRICES1 IN T HE U .S . J,11 , 1, / 1i I ,,2, I /11 II

i I ! !, I ./ 1, ,21 1I : / I,00 ' 0 0 ' lIl '- J O I I i I I I " 5 i2 k-[~~~~:~~ t ,I , I I~~IIL lNr' i ~ I :" , 0 1 \U J I I J .ill' ~ {Ill f f , 1 ! / - - " ' \ ,~1 ~ r r l j~rr ; r ~ I ' ~'I~dJ ,' I J ~ ~I'11 { !l I ~ W r lJh~'~I J ; ~ l U T t~1. t i l- J t J ' ! \ j I!, , , ,1 4 1 ) ) 1 1 7 - , I I , i ~ : ,:2 H~~ J J / 17 ff ! / ~~\ i11

/- THE 2 ,0 00 1" .A) . . . . :;Jr ;N t~"HOlf:S~U P f I I , a : IHDD:li~~--~~/ = : : : : I ~ L.'S. 1. ,6? ~ .OO, 1~ O n . 'f l J IU UG ;9O'--a".1 .1- ~..J'j....i_.... " ue e " iiiI . , 1 . l.it L,, ;r i~ f l ; n l"ij -.~ ' I I l i" ,. - -~--.~~~ ~

80 0

zo o

100

oo a

= _ JFigure 4-9 Dow Jones 30 Industrials (monthly high-low). (Source: CommodityResearch Bureau, New York.)

C ORR ECT IO NS IN A SH OR TT ER M T RE NDSwing Breakout StrategyIn order to showwhyworking with corrections can improve an invest-ment, a simple swing breakout system will be used as an example.Even though this strategy appears to have nothing to dowith ElliottorFibonacci, there is an underlying 3-waveconcept that can be closelyrelated to parts of Elliott's theory.

In a simple swing breakout system, a buy entry point occurs at aprevious swing high and a sel! occurs at the previous swing low.Thisconcept works beautifully as longas there are "regular" markets withbig swings, as shown in Figure 4-10. This method does not workwellduring sidewaysmarkets, shown in Figure 4-11.

".

7

5

63

. . S .U ) ' 84

2

Figure 4-10 Swing breakout system ina trendi ng market.

. ~.

In order to apply this method, parameters and rules must befully defined. Different swing sizes, stops, profit targets, trailingstops and re-entry rules can change the performance profile. Differ-ent rules a11'30pplywhen using intraday, daily, or weekly data.Swing Size. A swing high on the daily British pound chart is con-firmed after a decline of400 basis points occurs (exceeding the mini-mum swing value jn this example); a low is confirmed after asubsequent rally of400 points.Entry Point . A breakout above the previous swing high is a buy sig-nal; a fall belowthe previous swing lowis a sell signal.profit Target . A profit target established as the total amplitude ofswing] (the size of the previous swing in the entry direction) times

973 " v B "uy 5 Bu y a Sell

2 Sell 6 104 12

figure 4-11 Swing I1reakout system in a sidewards market.

56 WORKING WITH CORRECTIONS CORRECTIONS IN A SHORT-TERM TREND 57


38/94

.618(Figure 4-12). Once this profit target is reached, the current posi-tion is exited and noposition is entered until the next newentry signal.

Using a hypothetical example (Figure 4-12), the profit target iscalculated as:

57.20 - 50.10 = 7.10he total amplitude ofwave 1 isThe profit target is 0.618 timesthe size of wave 1,orThen, the profit target is

.618 X 7.10 = 4.3857.20 + 4.30 = 61.68

61.66 Profit- j - - - - - - - - - - - - - - - - - - - - - - - targetlevel

0.6185 7 . 2 Q L _C D

1.0I 50.10

Figure 4-12 The ratio 0.61 8 is used to calculate the profit target based onwave 1.

Figure 4-13 shows the strategy applied to the daily Britishpound over a six-month period.If the price reaches the profit target 61.68, the long position isliquidated.

The strategy ofworking with a profit target has both advantagesand disadvantages. If there is a 3-swing move followed by a trendchange, this strategy performs perfectly, capturing profits. But thereare two negatives: The price might miss the profit level completely,orwemight take profits onthe longposition only to see the price movemuch higher. It again depends on the investor's risk preference inmaking the choice ofwhich strategy to use.

0 : : ; : ; ~- - ': ; ; . ; i 'a : ' ; ': : ; .~ : I " :t I l ; , I I In i .1.. Ul l>; ;F. . . . . . . ;5 ;U t il lU,.. --...:" I .lm!IIWU.::;;'D ~ l : u ~ lG ~ l ~ f 51 ~ ~ : ! 1 ~ l nl c ~ t l ~ : ' i I a ~ 0 0 1 [: ~ ~ B

Figure 4-13 Dai ly Briti sh pound from 10-91 to 05-92. Buy and sell signals withprofit targets on a swing breakout strategy. (Source: System Writer Plus, OmegaResearch, Inc.)

Stop Loss. For a long position, the stop loss should be chosen as theprevious 2-day low and used as a trailing stop (use the opposite for asell signal).Tr ading S igna ls. The British pound example is intended to demonstratea very simple swing breakout strategy. It is not intended to be offeredas an overall profitable, complete trading system, but as a philosophicalapproach to howto imp

Documents

Robert Fischer - Fibonacci Applications & Strategies for Traders