Combining Technical and Fundamental Analysis

Combining Technical and Fundamental
Analysis
John Bollinger, CFA
President
Bollinger Capital Management, Inc.
Manhattan Beach, California
The efficient market hypothesis is flawed because investors do not have perfect
information and they do not always behave in rational ways. These inconsistencies in
investor behavior create exploitable opportunities and allow such tools as fundamental
analysis, technical analysis, behavioral analysis, and quantitative analysis to work.
Investors would be well served to explore all the tools available to them without
prejudice, especially in the present market environment because stocks are going through
a long-term consolidation period in which active management has a greater opportunity
to outperform buy-and-hold strategies.
echnical analysis has often been looked down

uponsometimes merely dismissed as useless
and sometimes treated with great disdain. A presentation on combining technical and fundamental
analysis would not have been well received a few
years ago. But times change. We are more than five
years into a century in which a buy-and-hold strategy has returned effectively zero, and people are
frustrated. In times of frustration, people open up
their minds and are more willing to re-examine the
tools (such as technical analysis) that are available
and how they might add value.
In this presentation, I will discuss combining
technical and fundamental analysis. Contrary to
what many people will admit, it is fairly prevalent, at
least in some minimal form, although perhaps
known by another name. Typically, the combination
of the two forms is called quantitative analysis;
sometimes, it is called behavioralism. Other times,
it is simply referred to as sneaking a peak at the chart
in the drawer. But no matter its name, it is an idea
whose time has come.
Efficient Market Hypothesis

For those who are not familiar with the efficient
market hypothesis (EMH), the strongest form says
that at any given point in time, security prices fully
reflect all available information. Thus, people outperform the market based on chance rather than skill.
60 CFA Institute Conference Proceedings
The EMH has three forms. The weak form asserts

that all past market data are fully reflected in securities prices. In other words, technical analysis is of no
use and I do not exist. The semistrong form asserts
that all publicly available information is fully
reflected in securities prices. In other words, fundamental analysis is of no help and many analysts do
not exist. The strong form asserts that all information
(public and private) is fully reflected in securities
prices. In other words, even insider information is of
no help and all those big houses and chalets in Aspen
do not exist.
The EMH is a nice theory. But in reality, the
market simply does not work that way because people are not the rational investors that market and
economic models assume. Emotions rule the roost.
They always have and probably always will.
In my view, the EMH is flawed at its core. Rather
than being an accurate assessment of information
content in securities prices, it is effectively an assessment of capitalization weighting.
So, why is the EMH an assessment of cap
weighting? A cap-weighted index is an incredibly
efficient relative-strength allocator. On each day,
each stock in the index that has done better than the
index receives a bit of additional weight, and each
stock that has underperformed the index that day
has its weight decremented in the index. In that
sense, a cap-weighted index is a portfolio strategy
2005, CFA Institute cfa pubs .org
Combining Technical and Fundamental Analysis
that involves no slippage, no trading costs, and no

emotions. It ruthlessly and relentlessly readjusts
itself each day favoring those issues that have done
better than it did that day and dishonoring and
decrementing those issues that have done worse.
That readjustment makes it an incredibly difficult
bogie to beat.
That said, I will look more closely at several of the
approaches people use when trying to beat that bogie.
Technical Analysis. Technical analysis has one
core tenet: The market is right. Technicians believe
that the most important source of information is the
market itself and will thus comb through price structure and price histories, relative-strength rankings,
volume analysis, and increasingly, analysis of investor behavior.
Technical analysis goes all the way back to the
1950s when Humphrey B. Neill started talking about
contrary investing. He was one the founders of Fidelity Financial Corporations chart work. Interestingly,
Fidelity, which is one of the most successful mutual
fund firms, still maintains a major technical presence.
Fidelity believes that technical analysis adds value to
its portfolio managers processes, and it believes that
a major commitment to technical analysis is entirely
consistent with its goal as a fund management firm
to provide superior returns for investors.
Technical analysis basically says that the EMH is
correct because most information is quickly
impounded into the market structure. But market
technicians do not take this statement to mean that
the market cannot be beat. Rather, they say that the
price record is what people ought to consult if they
want to understand anything. So, technicians simply
study the markets until they reveal their secrets.
Unfortunately, a large number of people in technical analysis believe in some kind of fantastic true
secret to the market structure. For example, some
technicians make investment decisions based on
Fibonacci sequences. In my view, such a practice is
charlatanism at its highest. But the fact that a few
people in the profession cleave to these ideas should
not dishonor the entire profession.
Fundamental Analysis. Those who use fundamental analysis believe that instead of the market
being right, the analyst is right. Typically, an analyst
builds a valuation framework, values the stock, and
then compares its price with the price in the
marketplace. If the predicted price is cheap compared with the market price, the analyst then views
the stock as overvalued. If the market price is less
than the predicted value, the analyst views the stock
as undervalued.
In essence, the division between technical analysis and fundamental analysis is over who is right.
Fundamental analysts believe that their analytical
tools and techniques are right. Market technicians
believe that the market is right and that they can add
value by somehow getting in tune with the market.
Thus, fundamental analysts typically say that the
EMH may be correct to a certain extent but that superior analysis can still provide an edge. So, fundamental analysts torture the books until they confess.
Quantitative Analysis. Prior to the mid-1960s,
quantitative analysis did not exist. Quantitative
analysis was simply a branch of technical analysis.
In the mid-1960s, quantitative analysis was born
with the surge in computer-processing power and
the ability to buy time on these early machines. An
analyst would buy time on the machine, punch his
or her program onto cards, and then go to the person
in the great white lab coat who guarded the sanctity
of the computer room.
The first quantitative analysts did no more than
codify in rigorous terms much of what technicians
had been doing for many, many years. With todays
computing power, quantitative analysts test to see
what works. Their view of the EMH is that it may be
right to some extent, but skillful numerical analysis
can provide an edge. They like to torture the data
until it confesses.
Behavioral Analysis. Behavioral analysis is
straightforward. It posits that people possess cognitive biases. Behavioral analysts are not so much concerned with what goes on in the markets. Instead, they
are concerned with the way people affect the market.
They view the EMH as flat out wrong because of its
core assumption of investor rationality. Behavioral
analysis asserts that there is no such thing as a rational
investor. Consequently, investors make systematic errors over time that result in exploitable opportunities.
I believe the behaviorists are essentially correct.
My personal proof is something called the options or
volatility smile. For those who are not familiar with
the options world, I will explain. The BlackScholes
model is the basic method used for deriving option
prices, and five variables go into the BlackScholes
model: stock price, strike price, time to expiration,
stock dividend, and volatility. Four of these five variables can be found by examination. But volatility
cannot; it must be estimated. Given the price of an
option, the model can be run backward and the volatility estimate pulled out in what is then called
implied volatility.
CFA Institute Conference Proceedings 61
Analyzing, Researching, and Valuing Equity Investments
An examination of the price structure of any

option and the corresponding implied volatilities
reveals something called the volatility smile. That is,
for a call option, a plot of various strike prices on the
X-axis and the corresponding implied volatilities on
the Y-axis shows that the implied volatility is higher
at lower strike prices and declines until the point
where the option is in the money. From that point,
the implied volatility begins increasing in conjunction with higher strike prices. So, plotting price versus implied volatility creates a graph that looks like
a smile.
The volatility smile, which options people have
known about forever, basically confirms the core
tenet of behavioral analysis. There are deep out-ofthe-money bets that the stock will trade well below
or well above the strike price. Those bets are like
lotteries and carry very, very small premiums paid
for very high, but unlikely, potential returns. People
make these bets for the same reason that they buy $1
lottery tickets: They are risking virtually nothing, and
the potential payoff is very large. This speculative
behavior at the extremes causes higher pricing than
one would expect, all else being equal, and this
behavior shows up in the volatility smile.
In sum, behaviorists study investors for weaknesses and inconsistencies. When they find weaknesses in judgment, they look for the corresponding
exploitable opportunities.
Rational Analysis. Back in the days when I was
working on my CFA charter, the mid-1980s, I was
active in the markets and the options world and active
as a technician. In those days, technical analysis was
talked down in the literature. I could not help but
wonder why that was the case. I did not understand
why something that could potentially add value was
consistently belittled. My belief was, and is, that these
different approaches should be combined into rational
analysis, where any tool that works is open for use.
It was about 1985 that I formalized this idea of
rational analysis. I came to think of it as the union of
the sets of technical and fundamental analyses. Since
then, I have come to realize that for rational analysis
to make sense, other toolsquantitative analysis and
behavioral analysismust be included. The union of
all these sets of tools results in the picture shown in
Figure 1. Note that the overlap is the area that I call
rational analysis.
Actually, I came late to the party. Chester W.
Keltner was a relatively famous commodity futures
analyst 50 years ago. In How to Make Money in Commodities, he wrote:
Actually, the best chance for success in commodity trading comes with a good knowledge of
both the two basic approaches to price
Figure 1.
Schematic of Rational Analysis
Fundamental Analysis
Technical
Analysis
Rational
Analysis
Quantitative
Analysis
Behavioral Analysis
forecastingthe price movement analysis (or

technical) approach, and the fundamental
analysis approach.1
None of the extra punctuation is mine. It is all Chester Keltners from 1961. Back even further than Keltner s work, in the article Relative Velocity
Statistics: Their Application in Portfolio Analysis,
H.M. Gartley wrote:
Price fluctuations, it would seem, should be considered as a factor in the valuation of a stock or a
portfolio of stocks. When an analyst has weighed
all the other variables he is accustomed to
employ, in judging a security, there is an addition
that might be called the technical factor. This
factor is derived from the interplay of supply and
demand, both at the time the appraisal is being
made, and historically. (p. 60)2
So, obviously, I was not the first one to the party.

If you were working on a car, would you limit
yourself to the tools in the red toolbox or the green
toolbox? Or would you simply use the best tools that
were available without regard to the color of the
toolbox that they came in? Why do we allow ourselves to be bound by a taxonomy that is not of our
own creation? Why do we allow ourselves to be ruled
by categories that have nothing to do with the economic value of the underlying systems that are contained within those categories? If you make that leap
with me, I think you will consider rational analysis.
1 Chester
W. Keltner, How to Make Money in Commodities, 3rd ed.

(Kansas City: Keltner Statistical Service, 1961).
2
H.M. Gartley, Relative Velocity Statistics: Their Application in
Portfolio Analysis, Financial Analysts Journal (April 1945):6064.
This article was reprinted in the 1995 January/February issue of
the Financial Analysts Journal and can be accessed online at
www.cfapubs.org.
At a presentation such as this, I find that if I ask

the audience who in the room has ever made an
investment in a publicly traded security without looking at the chart? I typically see only two or three
hands in a crowd of a hundred people. How times
have changed. Back in the 1960s and early 1970s,
people hid the charts in their desk drawers. By the
1980s, the charts moved out of the drawers and onto
desktops. By the late 1980s, the charts moved off the
desktops and onto computers. And by the 1990s,
everyone was looking at charts. In some way, we
have already accepted these ideas.
EMH Revisited
The real divide is the EMH. Those who believe that
the EMH is flawed ought to use the best tools available without prejudice to gain the greatest advantage.
Those who believe the EMH is correct should not be
at this conference (whose focus is the accounting tools
and techniques and such used in equity valuation).
In an efficient market, people cannot add value by
using these methods.
For 40 years, Eugene Fama, one of the intellectual
fathers of EMH, argued that financial markets are
highly efficient in reflecting the underlying value of
stocks. Recently, he surprised a group of economists
and business executives at a conference when he
presented a paper, co-authored with Kenneth French,
that conceded that poorly informed investors could
theoretically lead the market astray.3 When a luminary of tremendous education and tremendous
work, such as Eugene Fama, starts to talk about investors not being rational and about these core assumptions being flawed, we ought to pay attention.
My view is that the arguments are being conducted on the wrong terms. The question is not buy
and hold versus market timing. Instead, the question
is when to buy and hold versus when to market time.
During long consolidations after major expansions,
the name of the game is market timing. In such an
environment, market timing is the only way to add
value to a portfolio. During long expansions after
long consolidations, the name of the game truly is buy
and hold. Managers can add some value by focusing
on value, by focusing on relative strength, by focusing
on sector rotation, and so on, but the key is getting
aboard the train.
3
This paper has not yet been published. The article As Two Economists Debate Markets, The Tide Shifts by Jon E. Hilsenrath (Wall
Street Journal, 18 October 2004) describes the paper and the conference where it was presented. This article can be accessed online at
www.wku.edu/~bill.trainor/invest/Efficient%20Markets.htm.
Of course, beating the S&P 500 Index is hard when

the market is trending, but the market does not trend
all the time. One great fallacy of this business is that
one should focus on the markets bull market/bear
market cycles. Yes, the market goes through bull and
bear markets, but they are cyclical affairs and comparatively short term. Over the longer term is an alternating pattern of major expansions in the price structure
followed by major consolidations.
During expansions, the market is trending and a
buy-and-hold approach works well. In this environment, a manager adds value by avoiding mistakes.
This is the time when it pays to exercise rigor and get
rid of the stocks that are performing poorly while
focusing on the stocks that are doing well. During
consolidations, market timing is the key to earning
returns for clients. Managers add value through
stock, group, and sector selection.
Table 1 shows the past four cycles, which all ran
about 16 years. The market moved basically sideways
from 1934 through 1950 in the wake of the disruptions of the 1920s and 1930s. From 1950 through 1966,
the market trended upward following World War II.
From 1966 to 1982, the market moved sideways
again. Then from 1982 to 1998, the market moved
virtually straight up and included the longest uninterrupted rally ever. From 1998 through the present,
the market has been moving sideways and the S&P
500 is virtually where it was at its peak in 1998.
Table 1.
Market Timing vs. Buy and Hold
Start
Finish
Length (years)
Character
1934
1950
16
Consolidation
1950
1966
16
Trending
1966
1982
16
Consolidation
1982
1998
16
Trending
1998
Consolidation
Some may wonder why I have the latest expansion ending in 1998 instead of 2000. Figure 2 shows
that the Value Line Geometric Average peaked in
April 1998.4 The expansion from 1998 to 2000 was
limited to about only 150 large-cap stocks, which took
the cap-weighted averages through the roof.
An interesting point to note from a behavioral
point of view has to do with a phenomenon called
anchoring. Basically, the past three consolidations
4
The Value Line Geometric Average follows an equal-weighting

methodology as opposed to the market-cap-weighted methodology of the S&P 500. As such, the Value Line Geometric Average
approximates the performance of the median stock in the Value
Line Investment Survey.
Figure 2.
Value Line Geometric Average Index, January 1985May 2004
Index Value
550
April 1998
500
450
400
350
300
250
200
150
1/85
1/87
1/89
1/91
1/93
1/95
have been dominated by round numbers. During the

193450 consolidation, the Dow anchored around
100. From 1966 to 1982, Dow 1,000 was the anchor.
And from 1998 to the present, the Dow has been
anchored around 10,000. Interestingly, the same phenomenon holds for the S&P 500. The anchoring
behavior is not quite as clear as for the Dow, but the
S&P 500 has tended to anchor on 10, 100, and 1,000,
respectively, over the same periods.
During the major stock market expansions of
19501966 and 19821998, the market spent the first
couple of years just making up for lost time and
Figure 3.
1/97
1/99
1/01
1/03
getting away from being deeply undervalued. During roughly the past two-thirds of an expansion,
things start to heat up in anticipation of the forthcoming stronger economic reality. In consolidations, reality catches up to the markets. This is the time when
stock prices stop going up and earnings and economic activity play catch-up to stock prices. When
they finally catch up, the stage is set for the next cycle.
Figure 3 shows what this activity looks like in
terms of the S&P 500. The figure is in a logarithmic
scale to accurately portray the percentage changes
involved. Three consolidations are visible, with the
S&P 500 Index, 19282004
Index Value (log scale)

10,000
1,000
100
10
1
28
38
48
S&P 500 Index
58
68
78
88
98
Consolidation and Expansion Periods
last one starting in 1998 and being about seven years

old. The two major expansions are also visible. They
both lasted 16 years, and both the prior consolidations lasted roughly 16 years.
There is nothing magical about the 16 years. I
think it is happenstance. The lengths could have just
as easily been 10 years, 20 years, or anything in
between. In no way do I want to give the message that
16 is the magic number. The point is that the market
has expansions and consolidations. Then, within
those expansions and consolidations are what people
call bull and bear markets. During the expansion
phase, bull markets tend to run three to three-and-ahalf years and bear markets tend to run six to nine
months. During consolidations, the market spends
roughly two years in both bull and bear markets: Two
years going up, two years going down, and on net
going nowhere, which is where this idea of alternating
bull and bear markets comes from. But this idea of
alternating bull and bear markets is only evident in
the price structure on the cyclical level. It is not evident
in the price structure on the long-term level, where
one sees these expansions followed by consolidations.
Some might argue that these observations are an
artifact of modern day America because the United
States has been on a growth path for so long. I went
back and looked at the Cowles Index and some other
stock market surrogates into the mid-1800s. The patterns seem to hold, although I hesitate to rely too
much on recreations of old data because of the high
level of survivorship bias. Robert Shiller has data
going back to the 1800s that were supposedly
Figure 4.
adjusted for survivorship bias. These data are available on his website.5 But eliminating all the problems
is bound to be quite hard. So, ultimately, although I
am a little leery about using very old data because of
potential quality problems, these old data do tend to
support my hypothesis on long-term trending and
consolidating periods.
Figure 4 graphs S&P 500 earnings beginning in
1936, which was as far back as I was able to retrieve
data. The figure shows a simple linear regression line
(dark heavy line) fitted to the S&P 500 earnings numbers, which are again plotted on a logarithmic scale
to accurately depict percentage changes. I am really
struck by the relatively little variation around the
regression line. The fit of the regression line is very
good, and the standard error of the regression is very
small. The slope of this regression is 6.5 percent a
year, which is the long-haul rate of S&P 500 earnings
growth. The point that I would like to make with this
figure is that earnings grow regardless of the phase
of the market. Earnings grow whether the market is
in an expansion phase (as from 1950 to 1966) or in a
consolidation phase (as from 1966 to 1982), and stock
prices oscillate around earnings.
This earnings growth is the engine that drives
bull markets. People become pessimistic after the end
of a great bull market. By the time a consolidation is
over, they are no longer interested in investing. If you
tried to sell people stocks in 1982 or 1950, you probably found that they really did not want to talk to you.
5
This website can be accessed at www.econ.yale.edu/~shiller/.
S&P 500 Earnings, 19362004
Earnings (log scale)

100.0
10.0
1.0
0.1
36
46
56
66
76
86
96
Similarly, the circulation of Barrons goes down, and

so on and so forth. My own belief is that before this
current consolidation is over, some major structural
change will take place. TV channels like CNBC will
probably no longer be on the air. That kind of change
is typical of the investor disinterest that will prevail
before this consolidation is done.
Figure 5 shows the P/E for the S&P 500, again
going back to 1936; the median is 15.1. In the figure,
one can actually see the great cauldrons from which
the bull markets were born. In 1950, a P/E of 6 gave
rise to the great bull market running through 1966.
In 1982, a P/E of roughly 7 kicked off the most
recently ended bull market. The numbers in this
figure are not fantasy numbers. They are real. Note
that the market is nowhere near the low P/E that
investors typically see in a consolidation phase. In
fact, the market is not yet even back to normal.
As can be seen in Table 2, the market could be
back to normal relatively quickly, perhaps within
another year. As of 30 September 2004, the S&P 500
was at about 1,114. The next four quarter estimated
earnings are $71.55, given an estimated P/E of 15.58,
compared with the mean P/E of about 15 for 1936
2002. Remember that the long-term earnings growth
Figure 5.
rate for 19362002 was 6.5 percent. Based on that

information, the number of years for the P/E to go
from 15.6 down to 15 is one year.
Getting to undervalued, however, is a bit more
of a problem. As can be seen in Table 2, the first three
numbers stayed the sameroughly 1,100 for the S&P
500 price level, $71 for next four quarter estimated
earnings, and an estimated current P/E of roughly
15.6. Getting down to a P/E of 7, based on average
earnings growth of 6.5 percent, will take approximately 13 years.
A P/E of 7 corresponds roughly to the lowest 5
percent of observations from 1936 to present. Otherwise, a P/E of 7 is not particularly special. The important point is that the market will become undervalued
at some point in time. People will lose interest in
stocks, and the stage will be set for the next bull
market. I do not know whether that will be in two
years, four years, or eight years or what the P/E level
will be. But I doubt that the long-term psychology will
be different this cycle from what it has been in prior
cycles. We are human beings, and we are going to
play out these ideas again and again. We are not going
to play them out like a piece of music, note for note.
But we are going to play out the general themes.
S&P 500 P/E, 19362004
P/E
50
45
40
35
30
25
20
15
10
5
36
Table 2.
46
56
66
76
86
96
Earnings Growth Evidence
Years to Mean P/E

S&P 500 30/Sep/04 = 1114.58
Years to Low P/E

S&P 500 30/Sep/04 = 1114.58
Next four quarter estimated earnings = $71.55
Next four quarter estimated earnings = $71.55
Estimated P/E = 15.58
Estimated P/E = 15.58
Mean P/E (19362002) = 15
Average low P/E (19362002) = 7
Earnings growth (19362002) = 6.5%
Earnings growth (19362002) = 6.5%
Years to bring P/E to 15 = 1 (2005)
Years to bring P/E to 7 = 13 (2017)
Keys to the Locks

The current consolidation phase began in 1998 and
will likely last through the end of this decade, which
means tremendous opportunity for our profession.
Buy and hold will return little over the next several
years, but active management has the opportunity to
shine. By active management, I mean all forms of
active management, be it growth investing, value
investing, style rotation, or sector rotation. Each of
these disciplines will have its time. Probably the
superior way to approach this situation is to think
about what styles will perform well at any given
time. For example, small-cap stock as a style has been
the dominant return-producing idea over the past
couple of years.
A natural question is whether anybody is already
doing the kinds of things I have been talking about.
And the answer is yes. Charles M.C. Lee, in a piece
called Fusion Investing, combines fundamental
analysis and behavioral finance.6 He does not use the
term technical analysis, although I recognize what
he is doing as what I would call technical analysis. He
creates a four-stage stock life-cycle model very much
like Stan Weinsteins stage chart. He has a basing area,
an advancing phase, a top phase, and a declining
phase. What does he use to discriminate between
these phases? It is not price, value, or earnings; he
uses stock volume to discriminate between the various phases. It is a fascinating piece.
Another example comes from the Financial
Times, which maintains a knowledge website.7 It lists
a course called Fusion Analysis, which is taught by
John Palicka, CFA.8 He combines technical and fundamental techniques into quantitative screens.
To apply the ideas I have talked about in this
presentation, three approaches can be used.
Approach One. The first approach, which is the
most intuitively correct for people with a rigorous
academic and professional background, is to use fundamental analysis to determine what to own. Making
this decision means using stock models, and all of an
investors expertise comes into play. The next step is
to use technical analysis to determine when to own it.
Certainly more than one fundamental investor of the
value stripe has at some point suffered long and
mightily for owning a truly undervalued stock that
refused to go up for quite a while. By combining some
basic technical tools and techniques with a funda6
Charles M.C. Lee, Fusion Investing, in Equity Valuation in a

Global Context (Charlottesville, VA: AIMR, 2003):1523. This article
can be accessed at www.cfapubs.org.
7
This website can be accessed at www.ftknowledge.com/.
8
This course can be accessed at www.ftknowledge.com/courses/
tech_3002.html.
mental approach, investors can avoid that very long

waiting time and own those value stocks at a time
when they are more likely to produce rewards.
The most interesting work in recent years is a
book called What Works on Wall Street by James P.
OShaughnessy.9 Using the Compustat database, he
reduced survivorship bias down to a low roar. (I do
not believe that survivorship bias can be completely
eliminated, but he reduced it to an acceptable level.)
After going through numerous portfolios, his final
conclusion was that value and relative strength can
be combined to produce superior performance. To see
him go back and examine every issue in the Compustat database for a period of 70 years and come to
this conclusion was truly enlightening for me.
Approach Two. The second approach is to identify the group or the sector by using technical analysis
and then look into those groups or sectors by using
fundamental analysis to pick the stocks. Although
this approach can be very helpful, I have found its
reverse to be virtually impossible to do. That is, I find
group and sector fundamentals very tough to handle.
Some people, however, may be able to develop group
and sector fundamental forecasts and identify value
and then use technical analysis to enter and exit those
groups in a timely manner.
Approach Three. The third approach is already
being done a good bit, especially in recent years: Use
technical analysis to help manage the portfolio. That
is, use technical analysis to control risk by truncating
the left side of the return distribution with such tools
as stops. In addition, technical analysis can be used
to detect approaching good news as well as bad news.
Conclusion
Using technical analysis is no longer a strategy that
analysts need to be ashamed of. Indeed, it should be
seen as yet another tool in the analysts arsenal and
can be combined with fundamental analysis, behavioral finance, and quantitative analysisresulting in
rational analysis.
Figure 6 provides a good example of the usefulness of technical analysis. It is a chart of Enron Corporations stock price. Note that I have highlighted
one spot on the chart with a circle. That circle marks
where the first negative news announcement was
released by the company. Given the sustained downtrend in the stocks price, many technical analysis
techniques would have served as a filter or a warning
to get out of the stock long before the company
acknowledged anything.
9
James P. OShaughnessy, What Works on Wall Street, 2nd ed.

(Columbus, OH: McGraw-Hill, 1998).
Figure 6.
Enron Stock Price, June 2000January 2002
Price ($/share)
90
80
70
60
50
40
30
20
10
0
7/00
10/00
1/01
4/01
7/01
10/01
1/02
Q&A: Bollinger
Question and Answer Session

John Bollinger, CFA
Question: Can you recommend
any good books on technical
analysis?
Bollinger: The most important
early technical analysis book is
probably Richard Schabackers
Technical Analysis and Stock Market
Profits. Originally published in
1932, it has been reprinted by
Financial Times Prentice Hall. This
once very widely read book
focuses on chart patterns and the
psychology behind them. In many
ways, much of Schabackers writing is closely aligned with what we
know today as behavioral finance.
The next major milestone was
Technical Analysis of Stock Trends by
Robert D. Edwards and John
Magee. This is the book that several
generations of technicians cut their
teeth on, including many of
todays senior analysts. It is sometimes criticized as little more than
a rewrite of Schabacker, but it is
much more than that. This book
went through five editions with the
original authors in 1948, 1951, 1954,
1957, and 1962. In 1992, the New
York Institute of Finance prepared
a sixth edition, although John
Magee had passed on five years
earlier, and recently, W.H.C. Bassetti has updated it twice. From a
historical perspective, the classic
fifth edition is the most important.
In 1962, a retired General Electric Company statistician who had
taken up technical analysis, Arthur
A. Merrill, founded Merrill Analysis. He brought an especially high
degree of rigor to the process and
introduced statistical measures
and quantitative methods for technical tools and systems. An important example of his approach can
be seen in his groundbreaking
Behavior of Prices on Wall Street,
published by Analysis Press in
1965. See the second edition
revised in 1984.
Starting in 1978, Perry Kaufman took his book Commodity Trading Systems and Methods through
three editions with Wiley, shortening its name to Trading Systems and
Methods along the way. This series
was especially important because
Kaufman focused on systematic
trading methods and had a higher
degree of rigor than had been evident in the literature before, heralding a newer, more scientific
approach to technical analysis.
Next came the two primary
books read by todays technicians,
Technical Analysis Explained by Martin Pring and Technical Analysis of
the Futures Markets by John Murphy. The Pring book, currently in
its fourth edition, was originally
published in 1980. It focuses primarily on stocks, bonds, and the equity
markets. The Murphy book came
six years later, and by focusing on
the futures markets, it extended the
scope of the basic literature; the
second edition is called Technical
Analysis of the Financial Markets.
Together, these books form the core
reading for todays technicians.
Although many years younger
than Merrill, Martin Zweig was
one of his contemporaries and
blazed new trails in terms of rigorously specifying and testing trading approaches. He too was quite
innovative, good examples being
the measures he created using
option activity to gauge investor
sentiment and his linking monetary policy and equity prices. He
wrote two popular books but no
serious works, which is a shame
because he had a great deal to contribute beyond his oeuvre: Winning
on Wall Street and Winning with
New IRAs, both published by
Warner in 1986 and 1987. His work,
however, has been extended in
print by Ned Davis, Timothy
Hayes, and others, so all is not lost.
More recently, the horizon has

been extended by a Raytheon scientist, John Ehlers, who is interested in futures trading. He has
taken many of the tools he used so
successfully at work and translated
them to the markets. This is a prime
example of a new, more rigorous
approach to technical analysis that
has been gaining momentum.
Rocket Science for Traders was published by Wiley in 2001.
Of course, there are many
more authors and books, many of
them as important as or more so
than these selections, but this list
serves to sketch out the territory.
Anyone familiar with the works I
mentioned here will have a good
handle on the scope of technical
analysis as it stands today.
Question: Is mean reversion
behavioral or technical, and do you
believe in it?
Bollinger: I very much believe in
mean reversion. You see it on all
the longer-term charts. You see
mean reversion in the S&P 500
earnings. It keeps on swinging
above and below that trend line.
The problem is that people who
believe in mean reversion tend to
say that we are now N percent
away from some average, and
therefore, we ought to buy. That is
probably a pretty bad idea because
things can be on their way to zero
or they can be on their way to some
other level. I think we need to recognize that.
Question: Is there any evidence
that either fundamental analysis or
technical analysis works?
Bollinger: There is plenty of evidence that both fundamental and
technical tools work. And there is
plenty of evidence that behavioralists and quantitative proponents
are amassing.
If the question is whether some

technician can methodically yearin and year-out beat the S&P 500,
the answer is probably no. And
that is true for any fundamental
analyst. During those trending
phases, that relative-strength allocation engine that sits in the heart
of a cap-weighted average uses the
best strategy available. During the
consolidation phases, buy and
hold does not pay. So, a lot
depends on the phase of the market
that you are looking at.
Bollinger: The best use of

relative strength is a cap-weighted
index. It is the most efficient
relative-strength allocator you
want. Beyond that, I would add
that a mechanical approach would
probably work best for most investors because emotional responses
and knee-jerk reactions are among
investors greatest enemies.
Question: On the Enron chart,
are you saying that the circle represents the first sign that something
was wrong with the company?
Bollinger: It was the first admission by the company that something was wrong. When you have
a big decline like that, it ends up as
a pretty big news announcement. I
think the one bet that you can make
is that there will be more news
announcements to follow. You will
notice there was a bit of a relief
rally following the companys first
admission of problems. That is the
time to short a stock, not buy it,
especially when it occurs after a
prolonged decline.
Question: How is relative

strength best used?

Combining Technical and Fundamental Analysis

Uploaded by

Copyright:

Available Formats

Combining Technical and Fundamental Analysis

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Combining Technical and Fundamental Analysis

Uploaded by

Copyright:

Available Formats

What are the core tenets of technical analysis?

What are the core tenets of technical analysis?

What are some of the major works and authors that have contributed to technical analysis?

What are some of the major works and authors that have contributed to technical analysis?

Combining Technical and Fundamental

echnical analysis has often been looked down

Efficient Market Hypothesis

The EMH has three forms. The weak form asserts