Time the Markets

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page i

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page ii

Time the Markets

Using Technical Analysis to

Interpret Economic Data,

Revised Edition

Charles D. Kirkpatrick, II, CMT

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page iii

Vice President, Publisher: Tim MooreAssociate Publisher and Director of Marketing: Amy NeidlingerExecutive Editor: Jim BoydEditorial Assistant: Pamela BolandOperations Specialist: Jodi KemperSenior Marketing Manager: Julie PhiferAssistant Marketing Manager: Megan GraueCover Designer: Alan ClementsManaging Editor: Kristy HartProject Editors: Lori Lyons and Jess DeGabrieleCopy Editor: Cheri ClarkProofreader: Language Logistics, Christal WhiteIndexer: Erika MillenSenior Compositor: Gloria SchurickManufacturing Buyer: Dan Uhrig© 2012 by Charles D. Kirkpatrick, IIPearson Education, Inc.Publishing as FT PressUpper Saddle River, New Jersey 07458This book is sold with the understanding that neither the author nor the publisher isengaged in rendering legal, accounting, or other professional services or advice by publishing this book. Each individual situation is unique. Thus, if legal or financial adviceor other expert assistance is required in a specific situation, the services of a competent professional should be sought to ensure that the situation has been evaluated carefully andappropriately. The author and the publisher disclaim any liability, loss, or risk resultingdirectly or indirectly, from the use or application of any of the contents of this book.FT Press offers excellent discounts on this book when ordered in quantity for bulk purchases or special sales. For more information, please contact U.S. Corporate andGovernment Sales, 1-800-382-3419, corpsales@pearsontechgroup.com. For sales outsidethe U.S., please contact International Sales at international@pearson.com.Company and product names mentioned herein are the trademarks or registered trade-marks of their respective owners.All rights reserved. No part of this book may be reproduced, in any form or by any means,without permission in writing from the publisher.Printed in the United States of AmericaFirst Printing December 2011ISBN-10: 0-13-293193-1ISBN-13: 978-0-13-293193-9Pearson Education LTD.Pearson Education Australia PTY, Limited.Pearson Education Singapore, Pte. Ltd.Pearson Education North Asia, Ltd.Pearson Education Canada, Ltd.Pearson Educatión de Mexico, S.A. de C.V.Pearson Education—JapanPearson Education Malaysia, Pte. Ltd.Library of Congress Cataloging-in-Publication Data:Kirkpatrick, Charles D.

Time the markets : using technical analysis to interpret economic data / Charles D.Kirkpatrick II. — Rev. ed.

p. cm.Includes bibliographical references and index.ISBN 978-0-13-293193-9 (hbk. : alk. paper)

1. Technical analysis (Investment analysis) 2. Investment analysis. 3. Investments. I. Title.

HG4529.K5645 2012b332.63’2042—dc23

2011044607

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page iv

On a project such as this book, two requirementsare necessary: contemplative seclusion and the

basics of life: sleep, food, and something to pat. My dear wife, Ellie, and

our furry animals kept me secure, satisfied, and hidden. For that, I am intensely grateful. Ellie deflected potential disturbances, not to

mention providing timely sustenance; our goofygolden retriever, Posie, kept my feet warm during

the cold Maine winter nights; and Frisbee, our long-haired, reject cat from the Colorado

humane society, slept on my papers and acceptedmy occasional strokes of her fur.

To them I dedicate this book.

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page v

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page vi

Contents

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xiii

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

2 Why Time the Market, and Can It Be Done . . .7

3 Technical Analysis . . . . . . . . . . . . . . . . . . . . . .27

4 Systems Analysis . . . . . . . . . . . . . . . . . . . . . . .49

5 Corporate Indicators . . . . . . . . . . . . . . . . . . . .77

6 Economic Indicators . . . . . . . . . . . . . . . . . . .107

7 Monetary Indicators . . . . . . . . . . . . . . . . . . .127

8 Sentiment Indicators . . . . . . . . . . . . . . . . . . .145

9 Putting It Together . . . . . . . . . . . . . . . . . . . . .159

Appendix A System Parameters . . . . . . . . . . . . . .171

Appendix B Model Signals . . . . . . . . . . . . . . . . .175

Appendix C EasyLanguage Programs . . . . . . . . .179

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .183

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page vii

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page viii

Acknowledgments

In projects like this book, the Internet and one’s com-puter become your best friends. They provide informa-tion, complete complicated calculations, and remainalive 24 hours of the day. I am a night owl and preferthe quiet and calm of late nights and early mornings. Ihave a relatively small office at home that looks over theocean during the day but could be anywhere at night.To me, this is the perfect work environment.

Working late at night means that email becomes theprimary method of communication. Through such giveand take over the wires, many people familiar with thework I was doing assisted me. They helped considerably inthe book’s organization and application. Special thanks Iowe to Tom McClellan (www.mcoscillator.com) and JasonGoepfert (www.sentimenttrader.com) for their willingnessto provide their historical data. Tom also reviewed themanuscript for me and made numerous helpful suggestionsand changes to my fractured prose. Also helpful wereInvestors Intelligence (www.investorsintelligence.com), the Economist magazine (www.economist.com), and theCommodities Research Bureau (www.crbtrader.com).While I didn’t use all their data, they were very willing toassist and were especially generous with their knowledge ofeconomic statistics.

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page ix

x Time the Markets

The primary methodology used in the book is called“walk-forward optimization.” This technique is relativelynew to the securities business and technical analysis specif-ically, and is not widely known or understood. I, too, wasa novice when I began. I am especially grateful for helpwith it from Rob Hanna (quantifiableedges.blogspot.com),Bob Fulks of Pleasant Bay Capital Management, WouterOosthuizen of the Grail optimization system (now an inte-gral part of TradeStation analysis software), and BerkhardEichberger of the Diamond Back Testing with WalkForward Manager system from Professional SoftwareSolutions.

No book is the sole effort of the author. Many com-petent people at FT Press, most of whom I have nevermet, are responsible for the multiple aspects of publish-ing this book. I thank those with whom I have haddirect contact—Jim Boyd, Kristy Hart, Jess DeGabriele,Lori Lyons, and Cheri Clark—and those who are hid-den in the corporate structure somewhere for whateverexcellent work they do to get this project completed andon the shelves.

Finally, as the dedication explains, I could not havefinished this project without the help and encourage-ment from my dear wife, Ellie, and our furry householdcreatures.

Charlie KirkpatrickKittery, MaineOctober 24, 2011

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page x

About the Author

Charles D. Kirkpatrick, II, CMT, is president ofKirkpatrick & Company, Inc., a technical analysisresearch firm that publishes the Market Strategistinvestment newsletter. During his professional career, hewas an institutional salesman, technical analyst, portfo-lio manager, hedge fund general partner, securities trad-ing firm owner, option trader, floor trader, lecturer atuniversities, expert witness at security trials, small busi-ness owner, charitable foundation organizer and officer,and combat-decorated officer in Vietnam.

A past instructor in finance at the School of BusinessAdministration, Fort Lewis College, Durango, Colorado,and recently Adjunct Professor of Finance at BrandeisUniversity’s International Business School, he is the onlytwo-time winner of the Market Technicians Association’sprestigious Charles H. Dow Award for research in tech-nical analysis and winner of the Market TechniciansAssociation 2008 Annual Award for “outstanding contri-butions to the field of technical analysis.”

He is a Chartered Market Technician (CMT), a pastmember of the board of directors of the MarketTechnicians Association, and past editor of the Journalof Technical Analysis. He is a past member and vicepresident of the board of the Market TechniciansAssociation Educational Foundation and is currently a

xi

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xi

member of the American Association of ProfessionalTechnicians (AAPTA).

In addition to more than ten published articles onaspects of the stock and bond markets, he coauthoredTechnical Analysis: The Complete Resource for Finan-cial Market Technicians, the primary textbook for theCMT program and for university graduate courses intechnical analysis, and he authored Beat the Market, abook on relative strength stock selection. A graduate ofPhillips Exeter Academy, Harvard College (AB), and theWharton School of the University of Pennsylvania(MBA), he lives in Maine with his wife of almost 50years.

xii Time the Markets

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xii

Foreword“Every investor is a market timer. Some people buy whenthey have money and sell when they need money. Othersuse methods that are more sophisticated.”

—Marian McClellan, 1934–2003

My mother Marian taught me that lesson many yearsago, when I was first getting started as a stock marketanalyst. She had seen and heard a lot of good and badmarket “wisdom” over the years since she and myfather Sherman McClellan first created the McClellanOscillator and Summation Index back in 1969. Sincethat time, hundreds of thousands of people havebecome aware of the tools that they originated, and asmaller number than that have learned to use them suc-cessfully to help in their market timing.

The term “market timing” has taken on a negativeconnotation over the years, and that is unfortunate. The“buy-and-hold” community has sought to convince allof us that the key to investing success was to stay fullyinvested for the long run so that you don’t miss the bigup days that account for a lot of the gains. And you cancertainly find periods in history when that was a goodidea. But they conceal from you the fact that the biggestdown days are larger than the biggest up days and thatthe big down days tend to arrive in groups. The key toreal investing success is to make as much as possible

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xiii

when the market is going up and to lose as little as pos-sible when it is going down.

Sustained bull markets like the 1980s and 1990s aregreat when they appear. But there are more periods inhistory when being a “sheep” investor who just stayswith the flock has led to destruction of wealth. We arein such a period now. Baby Boomers are starting toretire and are no longer participating as much in theentrepreneurial economy like they did in the 1980s and1990s when Boomers were in their peak entrepreneurialyears. Now, Boomers are seeking to hold onto whatthey have rather than maximize their investing andentrepreneurial potential.

Boomers are hoping to sell their stock portfoliosand their McMansions to someone else, and in a fewyears the “echo-boomers” will be in a position toacquire those assets. But the “echo boom” peaked in1990, and those kids are still in college now. The echoboomers are neither ready nor able to buy yourMcMansion, let alone your bond portfolio.

We went through a similar period in the 1970s. TheUnited States had just come through more than 20 yearsof strong economic growth. But the people who were intheir peak entrepreneurial years during the 1970s hadbeen born in the 1930s and early 1940s—a time whenthe country and indeed the whole world was goingthrough the Great Depression. Birth rates dropped inthe 1930s because couples were afraid of having one

xiv Time the Markets

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xiv

more mouth to feed. So the kids who did not get bornin the 1930s also did not go on to become workers andentrepreneurs in the 1970s, which meant that both thestock market and the economy suffered as a result.

That did not stop the U.S. government from tryingto do things to “fix” the economy in the late 1960s and1970s. President Nixon tried wage and price controls,which were a colossal failure. The Federal Reserve keptinterest rates lower than the inflation rate in hopes ofstimulating growth, and sometimes that was successful.But it also led to huge inflation and wealth destruction.The ebb and flow of liquidity in the system at differenttimes created big waves up and down in the stock mar-ket. It was a great time for market timers and a lousytime for investors, just like the 1930s had been fourdecades earlier. And just like the 1890s had been fourdecades before that.

Now we are four decades forward from the 1970s,and once again, we have the Federal Reserve and thefederal government trying imaginative ways to fix theeconomy. So just like in previous periods, we are goingto see huge ebbs and flows of investing success anddestruction of wealth. The game that worked in the1980s and 1990s has changed; so if you are going toplay this new game, you will have to change your styleof play.

xvForeword

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xv

Timing Is Key

When you make up your mind what to buy, the onlycondition that is under your control is when you willpull the trigger. You don’t get to set the price; you haveto take whatever the market is offering. You can try tobuy a stock at a different price than what everyone elsethinks is the right price at that moment, but good luckconvincing anyone to sell it to you at less than what themarket sets.

And when you own an investment, the only ques-tion is whether you are going to hold onto it or sell it.If you are selling it, the essential question is “When?”You might say that you want to sell when it reaches acertain price or when it reaches some multiple of earn-ings, but that is not meaningful information to the mar-ket. The market wants you to say, “Sell now,” or “Don’tsell now.” Those are the only messages that the marketunderstands.

The people who say they do not time the market failto understand this essential reality. Everyone times themarket, whether he accepts that notion or not. The tim-ing of your investment decisions will have a huge effecton your success or lack thereof. So to say to yourself (oranyone else) that you are not a market timer is to saythat you willingly abandon the one factor that is in yourcontrol, and which is the key to your own success.

xvi Time the Markets

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xvi

xviiForeword

It is far better for investors to seek to maximize theirsuccess through mastering the factors within their con-trol, while also minimizing the effects of factors that arebeyond their control. To do otherwise is to be dishonestto oneself or to accept whatever the universe decides todo to you.

I have known Charlie Kirkpatrick for several years,and what I admire most about him is his willingness toshare useful information with others. Charlie loves toteach and to elevate the collective wisdom of the com-munity by sharing the great insights he has uncoveredover the years. Those of us who have been smartenough to open our ears when Charlie is talking havebenefited greatly.

In Time the Markets, Charlie leads us throughproven ways to time our investment decisions usingdata and facts that most of us can understand. Youwon’t have to learn to interpret tea leaves or pigentrails, to map star and planet positions, or programmathematically complex formulas into a “black box.”Just take the important data that are freely availablefrom government and other sources and learn how toread and understand what the changes in those datamean for the future of stock price movements.

Our country and our economy need the services ofinvestors who can appropriately add liquidity at theright time and take it away from the market at the right

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xvii

time as well. To those who can perform this great serv-ice, the market will give rewards in the form of a largeramount of money so that they can do those servicesagain in the future. But people who buy at tops and sellat bottoms are a hindrance to an efficient market, andthe market will punish them by diminishing their abilityto engage in such harmful behavior in the future.

You can choose to be in the “useful and therebyenriched” group and be a help to the market. Forinstructions in how to do that, read on and enjoy. Oryou can be a sheep. Sheep should close the book now.

Tom McClellanEditor, The McClellan Market Reportwww.mcoscillator.com

xviii Time the Markets

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xviii

Preface

The reasons for this revised edition of Time the Marketare twofold: First, the original edition had some com-promised data that slipped through the editing process.Second, and more importantly, the means of calculatingwalk-forward optimization became easier and morecomprehensive shortly after the original edition waspublished. This ease and increased accuracy, as well asgreater confidence in the outcomes, caused me to com-pletely revise all the indicator system calculations. Theresults are more accurate and have more credibility thanthose in the original edition. Indeed, you can be certainthat the system formulas and their outcomes are as up-to-date as possible. Those of you who wish to duplicatethe calculations should have little difficulty. I useTradeStation’s newly integrated walk-forward opti-mization programs. Using other programs may producedifferent results, but I have great confidence in the sys-tems that passed successfully through the TradeStationmethod.

If you have serious problems with duplicating the results, please let me know. My email address iskirkco@capecod.net.

Charlie KirkpatrickOctober 27, 2011

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xix

00_9780132931939_FM.qxd 11/17/11 10:46 AM Page xx

27

3

Technical Analysis

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 27

This chapter covers the technical analysis tech-niques that are used later in determining thesuccess or failure of market timing systems

based on economic data. They do not include the stan-dard chart pattern analysis, but instead they include theanalysis of trends using moving averages and othermethods.

Trends

As any chart of market prices will show you, priceshave a predilection toward traveling in trends. Thetrend, of course, can be upward or downward at vari-ous slopes or sideways. Most investors in trading mar-kets make money following the trend of an investmentprice. The fact that prices trend makes it possible tomake money. If prices were purely random in theirmovement, no one would profit. But people do profit,and very handsomely, because prices travel in trends.From a technical perspective, a trend is a directionalmovement in prices that remains in effect long enoughto be identified and still be playable. Not all trends lastlong enough to be recognized and then acted on.Profiting also depends on the investment horizon of theperson analyzing trends. If his outlook is for long-termtrends, day-to-day price motion is irrelevant. If his out-look is to swing trade over a few days, the long-termtrend is unimportant.

28 Time the Markets

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 28

Regardless of the trend length, prices do not followa straight line. Around the trend, prices tend to fluctu-ate. When that trend changes direction, it is first evi-dent in one of the fluctuations. However, not allfluctuations are changes in trend. They may be justcountertrend oscillations about the trend that willreturn to the direction of the trend.

The small vacillations around a trend sometimesmake the trend difficult to identify. Shorter trends areparts of longer trends. Though trends may be obviousin hindsight, ideally, we would like to spot a new trendright at its beginning and spot when the trend hasended. This ideal, however, never happens, except byluck. No magic indicator exists to spot precisely thebeginning and end of a trend. Looking at a graph ofprices, an analyst can spot many trends of varyinglength and magnitude, but such observations are obser-vations of history only. A trend must be recognizedearly and last long enough to profit. If you spot it tooearly, your chances of failure are greater; perhaps itwas just an aberration or a smaller, countertrend move,or perhaps it was a new trend but not long enough orlarge enough to profit. If you allow more time to provethat the trend exists, the chances of failure are less butpotential profit is lost when the price continues in thenew direction without your position. There is always atrade-off between potential risk and potential reward.This is why so much effort goes into accurately recog-nizing the beginnings and ends of trends.

293—Technical Analysis

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 29

Momentum

In the trading markets “momentum” is a word that iscommonly used to describe the rate at which pricetrends are changing. Classically, a price “trend” is aseries of prices that generally head in the same direction(up, down, or sideways). However, we know thatprices do not trend in one direction forever. When anychange in direction occurs, we say the prices changedmomentum. The directional change need not be areversal in direction. A trend change can just be a dif-ferent slope or rate of change.

Imagine a car traveling at 60 miles per hour. The 60miles per hour is its speed, or its travel “trend.” Shouldthe car slow down, we say it is “decelerating.” It is stilltraveling in the same forward direction but at a slowerspeed, and to get to that speed, it had to decelerate. Inmarkets, when the price trend is not rising as fast as itwas at an earlier point, we say it is losing momentum,or decelerating. In prices, losing momentum can even-tually result in a trend reversal. The car can stop andgo backward. Changes in momentum thus occur beforechanges in direction, just as changes in the car acceler-ation or deceleration precede changes in direction. Forthis reason, we want to study momentum. It leadstrends’ directional changes.

This is why price analysts so thoroughly studymomentum in markets. If they can detect a change in

30 Time the Markets

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 30

momentum, they might receive a clue as to how theprice trend will change direction in the future.Momentum is an early warning device in markets.

The traditional manner of measuring momentum isto calculate the change in prices from one period toanother. If the change is constant, the momentum isneither increasing nor decreasing. If the changedeclines, we receive a momentum warning that a pricedecline may be ahead. Conversely, when momentumincreases, we receive a warning that an advance may beahead. You should be somewhat careful in interpretingmomentum change, however. A change in momentumdoes not always bring a change in price direction. Amomentum change can occur when the price trendslope is increasing or decreasing but not necessarilyreversing.

Because prices are never rising or falling steadilybut have intermittent oscillations back and forth, youmust use a method that can measure momentum yetreduce the effects of the minor oscillations. Techni-cal analysts do this by using moving averages.

Moving Averages and

Moving Average Crossovers

Moving averages are one of the most useful methods ofidentifying and profiting from trends in prices or in anyother economic data. They are one of the oldest tools

313—Technical Analysis

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 31

used by technical analysts, dating back to 1901 withthe work of mathematician R. H. Hooker [1]. Movingaverages dampen out most of the fluctuations shorterthan the length of the moving average. A 40-day mov-ing average will reduce the effect of any fluctuations of40 days or less, for example. One-day fluctuations arealmost completely erased. The moving average reflectswhat occurred over the entire 40 days rather than just1 day. When a moving average changes direction, weknow that the trend represented by that moving aver-age has changed direction.

An average is the sum of a number of specific data,such as prices, divided by the number. A 20-day priceaverage is the sum of 20 days of prices divided by 20,the number of days. (A “moving average” is the aver-age calculation performed over successive periods andusually plotted on a chart for clarity.) A 20-day movingaverage, as shown in Figure 3.1, is a calculation of the20-day average over some succession of days. Whenplotted on a price chart, the moving average is usuallya smooth line that dampens the effects of the minor,sometimes erratic oscillations in the data. It thus repre-sents the trend through that data over the period of themoving average and disregards the clutter around it. Itis a measure of the trend and is useful for determiningwhen the trend is changing. A rising moving averageindicates a rising trend over the period of the movingaverage. A declining moving average indicates a declin-ing trend. If we calculate a rate of change in a moving

32 Time the Markets

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 32

average, we can see changes in the trend and thus thetrend momentum.

333—Technical Analysis

Created with TradeStation. ©TradeStation Technologies, Inc. All rights reserved.

FIGURE 3.1 Dow Jones Industrial Average with 20-day simple moving average (September 14–December 31, 2010)

The easiest type of moving average to understand iscalled a simple moving average, or SMA. Analysts alsouse other types of moving averages, such as the expo-nential, the linearly weighted, the Wilder, the geomet-ric, and the triangular. There are even methods thatwill vary the moving average length based on the his-toric volatility of the prices known as “adaptive” mov-ing averages. For our purposes, the results of theseesoteric calculations provide no extra advantage. Thesimple moving average is easy to construct and sufficesfor all your calculations.

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 33

The use of moving averages in investing has beenwidely documented. It is the reason for the success ofmany commodity traders, and academics have shownthat methods using moving averages demonstrate sta-tistical significance. Early studies of moving averagesas a timing method for stocks discounted their value.These studies used crossovers of prices and movingaverages, not crossovers of moving averages to movingaverages, and were statistically primitive. Brock,Lakonishok, and LeBaron (1992) [2] conducted thefirst study to show the validity of using moving averagecrossover rules, as well as trading range break rules.They found that moving average crossover signals gen-erate statistically significant stock market directionalsignals. Since then, using market data in other marketsand in other countries, additional studies have con-firmed much of their original academic work. We usesimilar methods ourselves when we analyze the datafor signals in later chapters.

We know that the markets have many differenttrends going at one time. There is the long, seculartrend, and then the intermediate-term trend, the short-term trend, and other trends above, below, and inbetween. We can construct a moving average of anylength, provided that the price information is available.Generally, shorter-period moving averages representshorter price trends, and longer moving averages repre-sent longer price trends.

34 Time the Markets

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 34

If we calculate more than one moving average overdifferent periods, we see the changes in the shortertrend versus the longer trend. Eventually the shortermoving average will cross over and under the longermoving average. These “crossovers” can be signals ofimpending change in price trend direction. Any systemdeveloped to use these crossovers is called a “movingaverage crossover system.” The unknown variables insuch a system are the lengths of the two moving aver-ages. We can prejudge what those lengths should be, orwe can optimize the data to see what lengths give themost reliable signals.

A longer-period length includes more data andmore information. Each specific data point becomesless important. A large change in specific data thus hasless influence on the longer moving average. However,if this large change in data is the beginning of a signif-icant change in trend, it takes longer for the trendchange to be recognized. The longer moving average isslower to pick up trend changes but less likely to indi-cate a trend change incorrectly from a short-term blipin the data.

Figure 3.2 shows two moving averages in the dailychart of the Dow Jones Industrial Average (DJIA). Theshorter-length moving average, 9 days, oscillatesaround the 18-day average and has a wider range. The9-day is the “faster” moving average, and the 18-day is the “slower” moving average. The shorter-length

353—Technical Analysis

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 35

moving average is always the faster average because itturns more quickly when a trend change occurs. It isless reliable as an indicator of trend changes, however.In Figure 3.2, notice how the 9-day moving average(dashed line) makes its troughs after the actual pricebottoms, and the 18-day moving average (solid line)makes its troughs even farther after the actual pricebottoms.

36 Time the Markets

Created with TradeStation. ©TradeStation Technologies, Inc. All rights reserved.

FIGURE 3.2 Two moving averages: 9-day and 18-day (DowJones Industrial: November 4, 2009–December 31, 2010)

The lag in turning, however, has an advantage.That is the advantage of surety of the signal. A changein direction of a moving average is more accurate thelonger the moving average period. A crossover of a fastmoving average and a slow moving average will tend tooccur near the turning point of the slow moving aver-age, and thus, while occurring long after the actual turn

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 36

in prices, it is more reliable as a signal. The conflictbetween accuracy and reliability is a recurring theme inany technical signal. Reliability reduces loss and is thusa preferable characteristic of any signaling system. Forthis reason, moving average crossover systems aremore commonly used for their reliability, even withtheir late signals.

A flat trend results in moving averages oscillatinghorizontally and crossovers not followed by directionalchange in prices. This causes “whipsaws” in signalswhereby a buy signal is followed by a sell signal at orbelow the buy signal price, and vice versa. This majorsignal fault with moving average crossover systemsoccurs only when the trends are flat and the traderloses money chasing fluctuating signals.

Figure 3.3 shows a flat period in CoreLaboratories’ share price, when the moving averagecrossovers gave false signals called whipsaws. It is thusimportant that the moving average period lengths belong enough to bypass any flat trends in the price.Because this is not always possible, moving averagecrossover systems have a high rate of false signals.Fortunately, the losses are quickly recovered by reversesignals. We can reduce these whipsaws with filters andother methods but never can eliminate them. On theother hand, the advantage of a moving averagecrossover system is that it will catch every major trendchange and “ride” that new trend to its termination.

373—Technical Analysis

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 37

As long as markets trend, the moving averagecrossover method, when properly applied, will catchthe major trends.

38 Time the Markets

Created with TradeStation. ©TradeStation Technologies, Inc. All rights reserved.

FIGURE 3.3 Simple moving average (SMA) crossovers caus-ing whipsaws in a flat trend (Core Laboratory common stock,daily: October 19–December 23, 2009) from Technical Analysis,page 281

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 38

Ratio of Price to a Moving Average

We can also detrend the data by subtracting it from, ordividing it by, the moving average. The resulting data isa portrayal of the fluctuations about the trend as it isrepresented by the moving average. Figure 3.4 showsthe DJIA with a 20-day moving average again.Following the price chart is another chart showing theratio of the closing price to that 20-day moving aver-age. You can see the oscillations around the trend moreclearly in this lower chart. The peaks and valleys in theratio chart show the periodicity of price oscillations.Sometimes these are regular, as in a harmonic cycle,and sometimes they are irregular and of little predictiveuse. In this instance, they are regular. The lows, forexample, occur roughly every 68 days.

393—Technical Analysis

Created with TradeStation. ©TradeStation Technologies, Inc. All rights reserved.

FIGURE 3.4 Dow Jones Industrial Average, 20-day movingaverage, and ratio of current price to the 20-day moving average, showing 68-day cycle period of lows (January 4–December 31, 2010)

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 39

Cycles

The stock market, and most other markets, has distinctcycles. Prices oscillate up and down around a trend.Sometimes these oscillations show regularity in theiroccurrence beyond pure chance. We call them “cycles,”but they are not cycles in the harmonic sense. They areconstant intervals between successive price tops or bot-toms. They are also controversial. Some think cyclesare imaginary, visions in the eyes of technical analysts;others discount them because their behavior is unex-plained. Whereas cycles such as the 68-day are obvi-ously difficult to justify, others are obvious and asregular as the sunrise each day.

The most obvious and easily explained are the sea-sonal cycles in agricultural commodities. The most pre-dominant cycle in the stock market is the four-yearcycle. This stock market cycle makes an important low roughly every four years. Wesley Mitchell(1874–1948), economics professor and founder of theNational Bureau of Economic Research (NBER), dis-covered it. He observed that the U.S. economy from1796 to 1923 suffered a recession approximately everyfour years. The stock market over the past 200 yearshas shown the same periodicity. Table 3.1 shows thecycle lows over the past 100 years and the averageinterval between lows.

40 Time the Markets

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 40

413—Technical Analysis

There are other cycles in the stock market, but the most important, and the one we are concerned with here, is the four-year cycle. It is often associatedwith the business cycle, and because it bottoms everyfour years, it is also called the “Presidential” cycle forthe interval between Presidential elections. I believe ithas nothing to do with the Presidential election becauseit also occurs in most other countries and especially inthose whose elections occur at intervals other than fouryears. It has also occurred for well over 150 years andbegan long before the U.S. became an economic super-power. It is likely due to a combination of businesscycle and investor memory, but both thoughts areunproven. Nevertheless, it exists and is a very impor-tant factor when analyzing the probability of imminentmarket declines.

Of course, the business cycle is not a cycle in theharmonic sense either. Instead, it is a wide fluctuationin business activity with an irregular periodicity thataverages four to five years. However, it does affectstock market prices and bond interest rates.

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 41

42 Time the MarketsTA

BLE

3.1

Four

-Yea

r C

ycle

in t

he D

ow Jo

nes

Indu

stria

l, 18

96–2

010

(Ada

pted

from

Bre

sser

t, 1

991)

[3]

Dat

e of

Low

Low

%

Dec

line

Dat

e of

Hig

h C

lose

Hig

h %

Adv

ance

M

onth

s M

onth

s M

onth

s C

lose

from

Hig

h C

lose

to H

igh

Low

to

Low

to

Hig

h to

to

Nex

t L

owH

igh

Nex

t L

owL

ow

Aug

ust

8, 1

896

28–3

1.2%

Apr

il 25

, 189

977

175.

0%49

.032

.017

.0Se

ptem

ber

24, 1

900

53–4

6.2%

June

17,

190

178

47.2

%38

.08.

929

.2N

ovem

ber

9, 1

903

42–4

8.5%

Janu

ary

19, 1

906

103

145.

2%48

.926

.722

.2N

ovem

ber

15, 1

907

53–2

7.7%

Nov

embe

r 19

, 190

910

190

.6%

47.0

24.5

22.5

Sept

embe

r 25

, 191

173

–43.

6%Se

ptem

ber

30, 1

912

9428

.8%

39.5

12.4

27.2

Dec

embe

r 24

, 191

453

–40.

0%N

ovem

ber

21, 1

916

110

107.

5%36

.423

.313

.1D

ecem

ber

19, 1

917

66–4

6.7%

Nov

embe

r 3,

191

912

081

.8%

44.8

22.8

22.0

Aug

ust

24, 1

921

64–1

6.7%

Febr

uary

11,

192

616

215

3.1%

56.0

54.4

1.6

Mar

ch 3

0, 1

926

135

–47.

8%Se

ptem

ber

3, 1

929

381

182.

2%44

.141

.82.

4N

ovem

ber

13, 1

929

199

–86.

1%A

pril

17, 1

930

294

47.7

%32

.35.

227

.1Ju

ly 8

, 193

241

–49.

0%M

arch

10,

193

719

437

3.2%

69.7

56.9

12.9

Mar

ch 3

1, 1

938

99–4

0.4%

Sept

embe

r 12

, 193

915

657

.6%

49.6

17.7

32.0

Apr

il 28

, 194

293

–23.

5%M

ay 2

6, 1

946

213

129.

0%54

.249

.64.

5O

ctob

er 9

, 194

616

3–1

6.1%

June

15,

194

819

318

.4%

32.6

20.5

12.1

June

13,

194

916

2–1

2.9%

Janu

ary

5, 1

953

294

81.5

%51

.843

.48.

4Se

ptem

ber

14, 1

953

256

–19.

5%A

pril

6, 1

956

522

103.

9%50

.031

.218

.8O

ctob

er 2

2, 1

957

420

–27.

1%D

ecem

ber

13, 1

961

735

75.0

%56

.950

.46.

5Ju

ne 2

6, 1

962

536

–25.

2%Fe

brua

ry 9

, 196

699

585

.6%

52.1

44.1

8.0

Oct

ober

7, 1

966

744

–35.

9%D

ecem

ber

3, 1

968

985

32.4

%44

.226

.318

.0

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 42

433—Technical Analysis

Dat

e of

Low

Low

%

Dec

line

Dat

e of

Hig

h C

lose

Hig

h %

Adv

ance

M

onth

s M

onth

s M

onth

s C

lose

from

Hig

h C

lose

to H

igh

Low

to

Low

to

Hig

h to

to

Nex

t L

owH

igh

Nex

t L

owL

ow

May

26,

197

063

1–4

5.1%

Janu

ary

11, 1

973

1052

66.7

%55

.232

.023

.1D

ecem

ber

6, 1

974

578

–26.

9%Se

ptem

ber

12, 1

976

1015

75.6

%39

.321

.517

.8Fe

brua

ry 2

8, 1

978

742

–24.

1%A

pril

27, 1

981

1024

38.0

%54

.238

.515

.7A

ugus

t 12

, 198

277

7–3

6.1%

Aug

ust

25, 1

987

2722

250.

3%63

.161

.31.

8O

ctob

er 1

9, 1

987

1739

–21.

2%Ju

ly 1

7, 1

990

3000

72.5

%36

.333

.42.

9O

ctob

er 1

1, 1

990

2365

–9.7

%Ja

nuar

y 31

, 199

439

7868

.2%

42.4

40.3

2.1

Apr

il 4,

199

435

93–1

8.5%

July

17,

199

893

3815

9.9%

54.0

52.2

1.8

Sept

embe

r 10

, 199

876

15–3

7.8%

Janu

ary

14, 2

000

1172

353

.9%

61.9

16.4

45.5

Oct

ober

10,

200

372

86–6

.6%

Mar

ch 4

, 200

510

941

50.2

%24

.517

.07.

4O

ctob

er 1

3, 2

005

1021

7–5

3.8%

Oct

ober

9, 2

007

1416

538

.6%

41.3

24.2

17.1

Mar

ch 5

, 200

965

47

Ave

rage

s–3

3.3%

99.6

%47

.232

.015

.2

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 43

Cycle Terminology

Harmonic cycles are composed of three measures:period, amplitude, and phase. Because market cyclesare not true cycles in the harmonic sense—otherwisethey would have been identified more precisely longago and would be easily recognized through standardharmonic mathematics such as Fourier analysis—wefind that the only consistent measure is that of“period.” This refers to the time it takes to progressthrough one complete cycle from bottom to top to bot-tom again. Amplitude in markets, the amount by whichprices rise from bottom to top, is not easily analyzedbecause it varies with the volatility of the market,which in turn is based on the emotions of the marketplayers. It is quantifiable but is not predictable. Phaseis the position of the cycle in relation to other cyclesand is not considered in markets. The only measure weare interested in then is the period—how long the cycleis and thus when is it due to bottom in the future.

It is best to measure stock market cycles from bot-tom to bottom because tops are generally rounded andbottoms are usually sharp Vs. This difference in config-uration seems to be due to their different psychologicalbackgrounds. Panic often accompanies bottoms, andpanic can come very quickly to the mass psyche. Thus,market bottoms tend to be sharp and completedquickly. On the other hand, greed is the most prevalent

44 Time the Markets

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 44

emotion at tops, but greed takes more time to develop.Thus, tops are rarely sharp spikes but more often arerounded and at times difficult to identify even in retro-spect. In economic data series, the differences in shapebetween tops and bottoms are less obvious. Althoughwe might use different length moving averages to catchthe tops and bottoms, in economic data it does notseem to make much difference. We therefore use thesame length moving averages to hunt for tops and bot-toms in economic data.

There are various ways to measure cycle periods.The easiest is to look at a ratio chart like that shown inFigure 3.4. This chart shows the ratio of the currentprice to its 20-day moving average. As the price oscil-lates around the moving average, we see definite peaksand valleys in the ratio. If these peaks and valleysappear to occur at relatively equal intervals, we likelyhave a cycle period in the data. In Figure 3.5, the pricechart is of the DJIA on a monthly basis with a 24-month moving average, and the lower graph shows theratio of the current price with its moving average. Thefour-year market cycle is readily apparent and markedwith vertical dashed lines. This is the major marketcycle in the stock market and the one that we shouldconcentrate on for market timing of investments.

453—Technical Analysis

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 45

46 Time the Markets

Source: Created with TradeStation. ©TradeStation Technologies,Inc. All rights reserved.

FIGURE 3.5 Ratio of monthly close to a 24-month movingaverage showing 4-year cycle lows (April 1986–December2010)

Notice in Figure 3.5 that the cycle is not perfect.Nothing is. The major declines in 1987 and 2008 didnot occur at the normal four-year interval. It turns outthey are part of a longer speculative cycle, but for ourpurposes, the four-year cycle assumption is not per-fectly accurate. For this reason, as you will see in thenext chapter, we use filters and stops to prevent ourbeing hurt by unexpected events. These methods willsignal us to leave the stock market despite what thefundamental and technical analysis suggests.

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 46

Conclusion

The principal difference between this book and othertechnical analysis books is that we are looking at meas-ures of momentum in economic data with the intent todiscover technical signals of long-term market pricechanges. In other words, when an economic series sys-tem gives a sell signal, it will apply not necessarily tothe economic data itself but to stock market prices. Itmay also signal an economic recession, but we are nowconcerned with profiting from the market direction,not the economy’s direction. The systems we create arefrom moving average crossovers of economic data.These crossovers will give us specific buy and sell sig-nals that we test using sophisticated walk-forwardoptimizing methods for reliability and predictability.The final market-timing model includes the best ofthese systems.

473—Technical Analysis

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 47

Endnotes

[1] www.mcoscil lator.com/learning_center/kb/market_history_and_background/who_first_came_up_with_moving_averages/.

[2] Brock, William, Josef Lakonishok, and BlakeLeBaron. 1992. “Simple Technical Trading Rulesand the Stochastic Properties of Stock Returns.”Journal of Finance 47:5, 1731–64.

[3] Bressert, Walter. 1991. The Power of Oscillators/Cycle Combinations. Tucson, AZ: Walter Bressertand Associates.

48 Time the Markets

03_9780132931939_ch03i.qxd 11/16/11 1:55 PM Page 48

183

I N D E X

four-year cycle in Dow JonesIndustrial, 40

harmonic cycles, 44periods, 44phases, 44tops, 45

Ddata, 58

categories, 58in-sample data, 65length of, 59-60out-of-sample data, 65reporting delay, 59stock market data, 75

declining moving averages, 32default spreads, 134-136Dennis, Richard, 54discretionary systems, 53disposable income, 120dividends, 104-105Dow 36,000 (Glassman), 23Dow Jones Industrial, four-year

cycle, 40downward trends, 10drawdowns, equity curves, 70

Eearnings, 102-104earnings yield, 82-83economic indicators

income, 119disposable income, 120initial claims for unemploy-

ment, 121-122unemployment rate,

120-121industry, 113-114

capacity utilization, 114-115

housing starts, 116-117industrial production, 115Purchasing Managers Index,

117-119prices. See prices, 108

AA-D line, 130advisory opinions, 147-149algorithmic systems, 53-55amplitude, cycles, 44ATR (average true range), 98ATR (average true range)

stop, 68

Bbonds

corporate Baa bond rate, 129long-term U.S. bond rate,

129-131three-month Treasury bill rate,

131-134business cycle, 41buy-and-hold investment

philosophy, 19-24

Ccapacity utilization, 114-115capital gains tax, avoiding

investment decisionsbased on, 17

capitalization-weighted compos-ite index, 19

categories of data, 58Census Bureau, housing

starts, 116complete liquidation, 16-17compounded interest rates of

return, 21Conference Board, 113consumer credit, 141-142Consumer Price Index, 109-110consumer sentiment, 149-150core inflation rate, 109corporate Baa bond rate, 129corporate data, 78corporate interest rates, 128credit, consumer credit, 141-142crude oil price, 111-112cycles, 40-41, 46

amplitudes, 44business cycles, 41

13_9780132931939_Index.qxd 11/17/11 10:36 AM Page 183

184 Index

Efficient Markets Hypothesis(EMH), 8

EMH (Efficient MarketsHypothesis), 8

equity curves, 70, 72-74stops, 74

excursion ratio, 97exit strategy, importance of, 11

FFederal Reserve Model, 79-80filtered moving average

systems, 50four-year cycles, Dow Jones

Industrial, 40fundamental analysis,

definition of, 2future stock market return

models, 78-79Federal Reserve Model, 79-80valuation ratios, 80-82

G-HGlassman, James, 23Goepfert, Jason, 153greed, 45

harmonic cycles, 44Henry, John, 54Hooker, R. H., 32housing starts, 116-117

I-J-Kin-sample data, 65income, 119

disposable income, 120initial claims for unemploy-

ment, 121-122summary of, 122unemployment rate, 120-121

indicators, necessary versusunnecessary indicators, 4

industrial production, 115“Industrial Production

and CapacityUtilization,” 114

industry, 113-114capacity utilization, 114-115housing starts, 116-117

industrial production, 115Purchasing Managers Index,

117-119inflation, 109initial claims for unemployment,

121-122Institute for Supply Management

(ISM), 117interest

compounded interest rates ofreturn, 21

short interest, 154-156interest rates, 128-129

corporate Baa bond rate, 129long-term U.S. bond rate,

129-131three-month Treasury bill rate,

131-134investment systems, 53investment timing, compared to

trading timing, 3Investors Intelligence Services,

147-149ISM (Institute for Supply

Management), 117

L-Mlaw of percentages, 17length of data, 59-60liquidation, 16-17long-term U.S. bond rate,

129-131

MAE (maximum adverse excursion), 73

managing risk, 9defining risk, 9-10market neutralizing, 12-15market timing

benefits of, 19, 22, 24case study: U.S. stock mar-

ket 1926-2004, 19-22complete liquidation, 16-17compounded interest rates

of return, 21definition of, 15law of percentages, 17traditional arguments

against, 22

13_9780132931939_Index.qxd 11/17/11 10:36 AM Page 184

185Index

systemic versus non-systemicrisk, 11-12

MAR ratio, 96margin debt, 146, 150-152market neutralizing, 12-15market risk. See risk managementmarket stops, optimizing systems,

67-69market timing, 8

benefits of, 19-24case study: U.S. stock market

1926-2004, 19-22complete liquidation, 16-17compounded interest rates of

return, 21definition of, 15EMH (Efficient Markets

Hypothesis) and, 8investment timing compared to

trading timing, 3law of percentages, 17traditional arguments

against, 22maximum adverse excursion

(MAE), 73MDD (maximum drawdown), 71Mitchell, Wesley, 40model parameters, optimization

and, 61-64models of future stock market

returns, 78-79Federal Reserve Model, 79-80valuation ratios, 80-82

moderate credit risk, 129momentum, 30-31

moving average crossovers, 31, 35-37

moving averages, 31-37monetary indicators, 128

interest rates, 128-129corporate Baa bond

rate, 129long-term U.S. bond rate,

129-131three-month Treasury bill

rate, 131-134spreads, 134

default spreads, 134-136

time spread (yield curve),136-138

money, 140consumer credit, 141-142money stock, 140-141

money indicators, money, 140consumer credit, 141-143money stock, 140-141

money stock, 140-141money supply M2, 140-141moving average crossovers, 31,

35-37moving averages

declining moving averages, 32filtered, 50momentum, 31, 35-37ratio of price, 39SMA (simple moving

average), 33whipsaws, 37

mutual fund cash percentage,153-154

mutual funds, Putnam VoyagerFund, 14

N-ONBER (National Bureau of

Economic Research), 40neutralizing strategy, 12-15nondiscretionary systems, 53nonsystemic risk, 11

objective functions, 63, 84oil, crude oil prices, 111-112optimization

model parameters and, 61, 64walk-forward optimization

methodfirst step, 84-87second step, 87-92third step, 92-97

systems, 60market stops, 67-69model parameters, 61, 64

out-of-sample data, 65

13_9780132931939_Index.qxd 11/17/11 10:36 AM Page 185

186 Index

Pparameters, 50, 90

for dual moving averagecrossover systems of stockprices, 164

model parameters, optimiza-tion and, 61, 64

winning systems, 161-162percentages, law of, 17periods, cycles, 44Pessimistic Return on Capital

(PROC), 85Pessimistic Return on Margin

(PROM), 85phases, cycles, 44position sizing, 55price trends, 30prices, 108-109

capitalization-weighted composite index, 19

Consumer Price Index, 109-110

crude oil price, 111-112Producer Price Index, 110-111summary of prices, 112

PROC (Pessimistic Return on Capital), 63, 85

Producer Price Index, 110-111profit percentage stops, 67profit percentage trailing

stops, 99profits, real returns, 9PROM (Pessimistic Return on

Margin), 85protective stops, 66, 100-101Purchasing Managers Index,

117-119Putnam Voyager Fund, 14

Q-RQuarterly Survey of Plant

Capacity, 115

ratio of price, moving averages, 39

ratiosexcursion ratio, 97MAR ratio, 96relative return ratio, 97

valuation ratios, 78real returns, 9recognizing trends, 29relative return ratio, 97reporting delays, data, 59risk management

defining risk, 9-10market neutralizing, 12-15market timing

benefits of, 19, 22, 24case study: U.S. stock mar-

ket 1926-2004, 19-22complete liquidation, 16-17compounded interest rates

of return, 21definition of, 15law of percentages, 17traditional arguments

against, 22systemic versus nonsystemic

risk, 11-12risk-control methods, 55

Sselling short, 51sentiment, 146sentiment indicators, 146-147

advisory opinion, 147-149consumer sentiment, 149-150margin debt, 150-152mutual fund cash percentage,

153-154short interest, 154-156

Shiller, Robert, 81short interest, 154-156short selling, 166SMA (simple moving

average), 33spreads, monetary indicators,

134default spreads, 134-136time spreads (yield curve),

136-138standard optimization, 64standard optimization

program, 62stock market data, 75stock markets, 163

13_9780132931939_Index.qxd 11/17/11 10:36 AM Page 186

187Index

stops, 55, 98ATR (average true range)

stops, 68equity curves, 74profit percentage stops, 67profit-percentage trailing

stops, 99protective stops, 66, 100-101trailing stops, 66, 98-102

summary of prices, 112system analysis, considerations

for, 52, 55-57systemic risk, 11systems, 50

equity curves, 70-74optimizing, 60

market stops, 67-69model parameters, 61, 64

parameters for dual movingaverage crossover systemsof stock prices, 164

parameters of winning systems,161-162

technical systems, 163-164variables, 51

Ttaxes, avoiding investment deci-

sions based on, 17technical analysis, definition

of, 2technical systems, 163-164Texas West Intermediate, 111three-month Treasury bill rate,

131-133“three steps and a stumble”

rule, 131time spreads (yield curve),

136-138timing models, creating, 165-169timing (markets), 8, 15

benefits of, 19-24case study: U.S. stock market

1926-2004, 19-22complete liquidation, 16-17compounded interest rates of

return, 21definition of, 15

EMH (Efficient MarketsHypothesis) and, 8

investment timing compared totrading timing, 3

law of percentages, 17traditional arguments

against, 22tops, cycles, 45trading timing, compared to

investment timing, 3trailing stops, 66, 98-102trend change, 30trends, 28-29

recognizing, 29“two tumbles and a jump”

rule, 132

U-VU.S. stock market 1926-2004

(case study), 19-22unemployment rate, 120-121

initial claims for unemploy-ment, 121-122

valuation ratios, 78-82Van Tharp Expectancy

Function, 63variables, 50-51Voyager Fund, 14

W-Xwalk-forward efficiency

(WFE), 91walk-forward method, 87-89walk-forward optimization, 6walk-forward optimization

method, 83-85first step, 84-87second step, 87-92third step, 92-97

WFE (walk-forward efficiency), 91

whipsaws, moving averages, 37winning systems, parameters,

161-162

Y-ZYardeni, Ed, 79yield curves, 136-138

13_9780132931939_Index.qxd 11/17/11 10:36 AM Page 187