INVESTORS MAKE BOTTOMS; TRADERS MAKE TOPS

AN ANALYSIS OF TECHNICAL INDICATORS THAT SHOW HOW THE ATTITUDES

AND ACTIVITIES OF STOCK MARKET PARTICIPANTS AID FORECASTING

PHILIP J. ROTH, CMT

MASTERS THESIS
DEPARTMENT OF ECONOMICS
RUTGERS, THE STATE UNIVERSITY OF NEW JERSEY
JUNE 2015

INVESTORS MAKE BOTTOMS; TRADERS MAKE TOPS

AN ANALYSIS OF TECHNICAL INDICATORS THAT SHOW HOW THE ATTITUDES
AND ACTIVITIES OF STOCK MARKET PARTICIPANTS AID FORECASTING
PREFACE
My intention is not to defend or deny the theories of Behavioral Finance. It has long
seemed a matter of certainty to technicians that the notion of Random Walk has been
refuted and that fundamental analysis is inadequate; hence, to us it was a matter of time
before academia would be searching for new explanations of stock price movements. Chan,
Jegadeeish, and Lakonishok wrote in 1999: What has caught the attention of many investors
recently is the predictability of price changes over intermediate horizons (three months to a
year). In the medium term, stock prices exhibit momentum (continuation in a price direction).
So, for these horizons, what goes up tends to keep rising and vice versa. 1 Some equity market
participants believe that stocks trend because information comes to the market place
incrementally or that investors only act on the information incrementally. However,
technicians are primarily concerned with the observation that stocks do trend. Prechter and
Parker wrote: In finance, uncertainty about valuations by other homogeneous agents induces
unconscious, non-rational herding, which follows endogenously regulated financial fluctuations.
This dynamic produces non-mean-reverting dynamism in financial markets, not equilibrium.2
It is my intention to show that many technical indicators are, indeed, behavioral.
Good news for technical analysis has come in the form of supportive academic studies
of those trend and momentum indicators. Such research goes back at least to 1989 when
Andrew Lo and Craig MacKinlay showed with a simplestatistical test that Stock Market Prices
Do Not Follow Random Walks.3 Subsequent research has found information (in the
statistical sense) in stock market trading volume ( Lo and Mamaysky, 2000)4, supportg and
resistanceg, including stickiness at round numbers and previous significant new highs and
lows (Osler, 2000, 2001 and 2003; Mizrach and Weerts, 2007)5, moving average trading rules
(Papailias and Tomakos, 2011)6, patterns and trends (Lo, Mamaysky, and Wang, 2000; Osler and
Chang, 1995; Savin, Weller, and Zvingelis, 2007; Weller, Friesen, and Dunham, 2007)7,
momentum indicators, specifically RSIg and MACDg (Lachhwani, Hitendra, and Khodiyar, 2013)8,
and Elliott Waveg analysis (Magazzino, Mele, and Prisco, 2012)9. Magazzino, Mele, and Prisco
wrote: The major findings of our analysis show that, in the case of turbulent financial markets
(such as we have been experiencing) technical analysis and the Elliotts theory adequately
reflect the realities of the financial markets

Professors Lo, Mamaysky, and Wang wrote after testing ten technical patterns: While
human judgment is still superior to most computational algorithms in the area of visual pattern
recognition, recent advances in statistical learning theory have had successful applications in
fingerprint identification, handwriting analysis, and face recognition. Technical analysis may
well be the next frontier for such methods. for NYSE/AMEX stocks, 5 of the 10 patterns HS,
BBOT, RTOP, RBOT, and DTOP, yield statistically significant test results10
Technical Analysisg is based on three premises, which those studies support. They are:
(1) Information is absorbed incrementally in the market place; it is not efficient. Hence, (2)
stocks trend; it is not a random walk. And, (3) those trends are, in the words of Robert
Prechter: probabilistically determinable. After writing his book of conversations with
technical analysts (Heretics of Finance, 2010)11, MIT professor Andrew Lo said in a presentation
to the Market Technicians Association he had a hard time trying to quantify the technical
indicators, but he was convinced traders were seeing something useful in the charts and he
believed the computers may eventually figure it out. Its like facial or pattern recognition.
Humans have little trouble distinguishing a cat from a dog, but teaching a computer to do that
has proven to be very difficult (albeit, probably not impossible) .
Long before fundamental analysis was codified and no one had heard of quantitative
analysis, excessive valuations were noted in markets. In 1841 Charles Mackay wrote about the
South-Sea Bubble in London, which occurred in 1711, and Tulip Mania in Holland, which
occurred in 163714. Those, of course, were examples of irrational exuberance, centuries
before the 1990s, which prompted Alan Greenspan to coin that phrase 15 and Robert Shiller to
write a book with that title16.
Technicians have called behavioral indicators sentiment or psychological indicators,
and supply/demand or money flow indicators. I will not be concerned with such aspects of
Behavioral Finance as the endowment effect, loss aversion, anchoring, etc.12 I will only be
concerned with optimism and pessimism on equities per se. In other words, I am interested in
who is bullish or bearish, and how bullish the bulls are and how bearish the bears are, not why
they may be bullish or bearish. I will refer to stock most of the time in this discussion, but the
theory applies to anything that trades continuously in an open market, such as bonds,
commodities, options, futures, and currencies.

INTRODUCTION
By now most equity market participants have learned about the notions of overpopularity and over-pessimism. They have read too much bullishness after big advances.
One such example is the book, Dow Jones 40,000 by David Elias. This book was published in
June 1999, after nearly two decades of rising stock prices, and only six to nine months before
the end of the secular bull market, and before ten years of no price progress, interspersed with
two lengthy, deep bear markets. People have read about too much pessimism after big
declines. An example was the famous (infamous?) Death of Equities cover of Business Week
magazine on August 13, 1979.17 The bearish article followed a decade and a half of no price
progress in stocks. In fact, it appeared near the lows for the next three years and was followed
by the great secular bull market of the 1980s and 1990s. However, those are anecdotal
examples and they imply everyone (I exaggerate, of course; most participants is what I
mean) is too bullish at tops and too bearish at bottoms. This is not the case, nor can it be the
case. Someone is selling the stock that traders cant get enough of late in bull markets and
someone is buying the stock traders are disgorging indiscriminately late in bear markets.
My sentiment (i.e., behavioral) indicators are of two types: those that deal with equity
market participants who have short term horizons and those that deal with equity market
participants who have long term horizons.18 I have found that investors make bottoms and
traders make tops. That is not to say investors are smart and traders are dumb. There is such
a thing as rational exuberance. Both savvy investors and savvy traders can and do show good
results. But they are motivated by different things and dominate the buying and selling in the
market at different times. Hence, my sentiment analysis is not about contrary opinion, unless
of course, we make it clear who to be contrary to and when to be contrary.
During lengthy trends most equity market participants get the message and are on the
right side much of the time. However, near turning points we usually find traders on the wrong
side for a while and investors on the right side. In fact, I define a bottom in the market to be
that period when stocks are stabilizing after a lengthy decline and investors are buying stocks
from traders, and I define a top in the market to be that period when stocks are faltering after a
lengthy advance and investors are selling stocks to traders. Of course, there is no way to know
what all investors and what all traders are doing, but we do have indicators that suggest what
the majority are doing. Why should it be that traders are often on the wrong side at turning
points and investors are usually on the right side?
Investors are motivated by price and value. As a result, they get more bullish and more
active on the buy side when prices are declining and when stocks are low. The opposite is
also true; they get more bearish and more active on the sell side when prices are rising and
stocks are high. Traders are usually not concerned with value; they are motivated by trends.
4

When stocks have been rising traders get more bullish; when stocks have been declining
traders get more bearish. Traders (I am including professional buy-side and sell-side traders, as
well as individuals trading for their own accounts) cannot recognize a final high price as it
occurs. In fact, technicians go to great lengths to emphasize that a topg, a reversal of an
advance, is not a point in time; it is a process that entails a series of weakening rallies and
growing declines. As the process of a top, a distribution patterng, develops, traders will begin
to get the message and move from a bullish posture to a bearish one. Investors who have been
bearish, who have been supplying stock, will continue to do so, because prices are still high.
The supply/demand balance will be tipping to the negative side. A bottomg, of course, is the
opposite process. Traders, who have been bearish because of declining stock prices, need
some time to recognize a stabilizationg pattern becoming a base, an accumulationg pattern.
Investors who have been bullish, who have been accumulating stocks, will continue to do so,
because prices are low. The supply/demand balance will be tipping to the positive side.
Therefore, importantly, my sentiment indicators distinguish investor attitudes and
actions from trader attitudes and actions. Attitude indicators include polls and anecdotal
information, i.e., what participants are saying. Action indicators are transactional measures,
i.e., what participants are doing. Together, indicators of attitudes and actions form the
framework for what market technicians call Sentiment Analysis, which I believe is a key
element in behavioral economics or behavioral finance. I will start with the longest term
indicators and end with the shortest term. While we have all heard many times that technical
analysis is arcane, or worse, divination, I believe sentiment analysis is reasonable and
obvious to knowledgeable market participants. When I have asked my students, Whose
advice and actions would you rather follow, Warren Buffet or the famous hedge fund manager
they just saw on CNBC, they always picked Warren, the consummate successful long term
investor. And most of them took my course because they wanted to learn how to become a
good trader.
Sentiment indicators cannot be used in isolation from market movements. Sentiment
indicators are useful only with respect to the trend. During much of a trend, up or down, most
people get the message and are on the right side. Too much (trader) optimism is only a
negative when price progress is struggling after an advance (and investors are supplying stock).
The opposite is also true. If you bought the first time you heard Theres too much pessimism
in a bear market, you would probably be buying in the middle of the decline. The late money
manager, Martin Zweig, was known for saying, Dont fight market momentum and dont fight
the Fed, simple but excellent advice. We would add Dont ignore the buying and selling
proclivities of investors and traders, especially when prices have stalled after lengthy rises or
falls.

The utility of the available transactional sentiment measures varies and changes with
market conditions. For example, in the 1950s and 1960s a useful indicator of public sentiment
was oddlot activity. The public was much more dominant in the markets and the average price
of stocks was much higher (i.e., a round lot position would have been too expensive). As a
result, odd-lot short sales reflected public trading sentiment very well: a lot of short selling late
in market declines and in bottom areas, and little shorting late in market advances and in top
areas. Now odd-lot activity is more the bailiwick of professional index-fund-oriented
professionals, who often need to use odd-lots to rebalance portfolios. Hence, such activity no
longer appears to be a very useful sentiment indicator. Some transactional indicators can be
distorted by changes in market activity. For example, round-lot short sales indicators are
affected by hedge shorting and arbitrage shorting. In this paper, I have chosen not to use oddlot activity and round-lot shorting at all.

PART ONE
VERY LONG TERM (I.E., SECULAR TREND) INDICATORS
I have found that different sentiment and supply-demand indicators are useful for
different time horizons. Indeed, the very long term indicators are useless for short term
horizons. The indicators we follow are useful for the very long term (the secular trendg, the
trend that encompasses two or more economic cycles), the long termg , (the cyclical trend; one
complete bull-bear cycle), and the medium termg, (intermediate term, usually one leg of a bull
or bear cycle, often three to six months). Sentiment indicators are usually not helpful for the
short term (near termg, from a few weeks to a few months) and are unlikely to be useful at all
for the very short term, such as day trading. On rare occasions, however, very extreme
readings in medium term indicators may be helpful for the short term.
1. Our first indicator is the most basic one, the Net Change in Supply of Stock. 19 The
data, derived from the Federal Reserve Flow of Funds series (FRB FOF, table F.4, line
12), is quarterly data, at a seasonally-adjusted annual rate (SAAR)20. The data is
released with about two-and-a-half month delay. With just four data points a year
and a long lag it is obviously only useful for a very long term perspective. What does
it tell us? When the series is on the plus side, it means new equity financing (IPOs
plus new stock sold by existing public companies) exceeds corporate buybacks. Who
is perhaps the most important long term investor? The corporation itself.
Corporate activity in its own stock is by its nature a very long indicator. When a
corporation comes into the market to buy back its own stock, it is almost a
permanent reduction in supply. Corporations action in their own stocks is
investment activity. Chart 1 shows that the market tends to do relatively well when
buybacks exceed new offerings. The supply of stock is decreasing. The second half
of the 1990s and 2004-2007 are great examples. The market tends to do relatively
poorly when new equity financing exceeds buybacks. The supply of stock is
increasing. 2001-2002 is a good example. Net corporate investment demand is
bullish and net corporate supply is bearish. It is Economics 101. Keep the demand
for something fixed and reduce its supply. What happens to the price? It rises.
Keep the demand for something fixed and increase its supply. What happens to its
price? It falls.

All the charts in this paper have been produced from my personal data, made available to the Market
Technicians Association Educational Foundation and its educational programs.

The Flow of Funds reports are also useful in determining whether individuals and
institutions own a lot of stock or a little bit of stock, compared to other financial
assets. Chart 2 shows the value of financial assets for Households20 and the equity
percent (equities plus mutual funds) of those financial assets (FRB FOF, Series L. 100,
lines 1, 18, and 19), compared to the S&P 500. The Federal Reserves Household series
is a proxy for the public. It is not exactly the public, since foundations are included in
the totals; nevertheless, it is a useful gauge. Note that the secular bull market in the
1980s-1990s started with a low equity percentage (13%-15%) and ended with a
high percentage (30%-33%). The bear cycles of 2000-2002 and 2007-2009 resulted in
retracements in the equity percentage of no more than half the rise and the post-2009
bull market has taken the percentage back close to the highs. Assuming that
Households are long term investors (much of the public financial assets are held in 401K
and IRA retirement accounts), we interpret the equity percent series to be a limiting
factor to equity potential. In the 1980s-1990s the great stock market rise was a
function of a lot of new money coming into stocks and money flowing from fixed income
investments into stocks. Therefore, this class of investors is probably less of a demand
force: new cash flow is still a factor, but asset switching (from fixed income investment
8

to equities) is probably much less of a factor. This indicator belies the popular notion
that the public has been eschewing stock.

2. Institutional categories give a similar message. Chart 3 is the series for State & Local
Government Employee Retirement Funds20 (FRB FOF, L. 118, lines 1, 13, and 14). It
shows that the equity percent (equities plus mutual funds) of financial assets soared
from below 20% to above 80% in the 1980s-1990s and that percentage stands at
52% now. We draw the same conclusion as for Households, namely that a great
secular bull seems unlikely with this class of investors never having shown a
complete washout from a high percentage of financial assets in equities. The
series for Private Pension Funds (FRB FOF, L.117, lines 1, 12, and 13), Life Insurance
Companies (FRB FOF, L.115, lines 1, 13, and 14), and Property-Casualty Insurance
Companies (FRB FOF L.114, lines 1, 12, and 13) look similar. This is not to say stocks
cant have cyclical bull markets with such a background, but a secular rise like the
1980s-1990s seem unlikely, with important classes of investors already highly
committed to equities.21

Since there are only 141 observations and the data lag by up to two-and-a-half months,
testing revealed no predictive power for the total financial asset series. The equity percent
would be predictive if there were no lag. The equity percent predicts the next quarter
returns (the coefficient for chgep = 0.01897, significant at the 1% level). See appendix I.

10

PART TWO
LONG TERM (I.E., CYCLICAL TREND) INDICATORS
A. INTERNATIONAL FLOWS
Indicators in Part One dealt with the change in total supply of stock and equity
allocations, indicators useful for secularg trends. In Part Two I will discuss measures that deal
with flows and valuations, indicators useful for the cyclicalg trend, the long termg trend. The
long term trend is the trend associated with the economic cycle, usually four-to-five years. The
first series of three charts deal with international flows. The data come from the U.S Foreign
Activity Report from SIFMA, the Securities Industry and Financial Markets Association22. SIFMA
refers to U.S. Investors and Foreign Investors. Those so-called investors act like traders,
however. The data show the participants seem more motivated by trend, than value. The
biggest net buying has occurred late in uptrends, while actual net selling has been seen near
major market bottoms. It seems that the further participants are removed from markets the
more likely they are motivated by the previous market action, i.e., they are more likely to be
trend followers than value seekers.
The first chart below (Chart 4) is net foreign buying of U.S. equities. Foreigners are
more trading-oriented in the U.S. market than U.S. participants are in the foreign markets
(Chart 5 below). Foreigners barely did any buying in the U.S. 1980s-1990s secular bull market
until the last couple of years. They were sellers near the market lows in 2002. Foreign net
buying has been erratic since the 2007 market top. The spates of selling in recent years during
the post-first-quarter-2009 bull market by these trading-oriented participants is one argument
for extending the life of the rise. That is, we expect to see traders very optimistic and,
accordingly, sustained buying late in a lengthy advance.

11

The next chart (Chart 5) shows U.S. buying of foreign stocks for the last 34 years. Note
that U.S. net buying was virtually non-existent in the 1980s bull market and, except between
mid 1998 and mid 1999 (which encompassed a sizable market correction), buying was greatest
after the mid 1990s. That sustained net buying was a record at the time. Buying and selling
were in balance in the 2000-2002 bear market, with the biggest selling occurring near the lows
in the third quarter of 2002. Post-2002, the figures show the biggest buying just before the
market highs in 2007 (the four quarters from the fourth quarter of 2006 through the third
quarter of 2007) and record selling in the fourth quarter of 2008 near the market lows. So far,
in the current up cycle the biggest buying was in the first quarter of 2013, after nearly four
years of market rise, and the second quarter 2013 net buying was the fifth largest ever.

12

The last chart (Chart 6) in this section is the most interesting of the three. It shows net
purchases of U.S. stocks by foreigners minus net purchases of foreign stocks by U.S.
participants, Net International Flows into the U.S. Market. Since foreign buying and selling
the U.S. market dwarfs U.S. activity in the foreign markets, this chart has the same shape as
Chart 4, but it gives a clearer picture of excess. It shows sizable net outflows from the U.S.
market until 1997 and record (at the time) net inflows just prior to the market top in 2000. It
shows net outflows at the market bottom in 2003 and little net inflows until the 2007 market
top. There have been no sustained inflows in the post-first-quarter 2009 bull market and there
were record outflows in the first half of 2013. When we compare international flows in the last
decade to previous decades it seems clear that traders (i.e., trend followers), especially U.S.
participants, are dominating the flows and investors (i.e., longer-term, value-motivated
participants) are reticent. I interpret that to mean we should expect the market to look more
like the post-2000 experience, rather than the 1980s-1990s experience. Bigger bulls and
bigger bears occur when traders produce the advances and declines, and investors are not very
active. That is, little investment selling in the face of trader demand and little investment
buying in the face of trader liquidation results in bigger swings.

13

This quarterly data, like the Flow-of-Funds-derived indicators, did not lend itself
to statistical testing. There are not enough data points. Nevertheless, I stand by the notion
that international flows do provide useful insight into equity demand (see above).

14

B. NEW SUPPLY
The very first chart was the Net Change in the Supply of Stock (see above). The data is
quarterly with a lag of two-and-a-half months. It would certainly be useful to have more
current information. The New Equity Financing series, while just the supply half of the Net
Change in Supply of Stock, is available monthly with less than a two-week lag, from Source
Media (www.sourcemedia.com). New Equity Financing includes initial public offerings (IPOs),
as well as new stock sold by existing public corporation. Chart 7 below shows the monthly total
and a moving 12-month total, compared to the S&P 500. The 12-month total mirrors the S&P
500, answering the question, When do most IPOs come out, at bottoms or tops? The
monthly totals almost always peak just before market peaks and show near zero or zero
readings at market bottoms. There is a secular rise in the data and it is not normalized because
I want the magnitudes to be clear. When prices are elevated corporations (market participants
with long term horizons) want to sell new stock and are able to do it because the outside
investors and traders have increased appetites for stock. I note that the longer an advance
persists, the more the demand comes from speculative sources (i.e., traders, short-termoriented participants. See Margin Debt below). After protracted market declines
corporations do not want to sell new stock (i.e., they perceive their stock to be too cheap)
and speculators are not inclined to buy down-trending stocks. Note the low levels of new
equity financing in 1990, 1998, 2001-2002, and 2008-2009, with most of the lowest readings
occurring at or near the ends of major declines. Even 1994, which was essentially a sideways
correction resulted in near zero readings near the end. Note the very high readings (i.e., the
peaks in the 12-month total) in mid 1983, mid 1987, early 1994, mid 1998, and early 2000. The
only exception of sizable new equity financing occurring during a major decline and early in a
major advance was during the 2007-2009 bear market and in the post-2009 bull market; this
was because of the major recapitalization in the banking industry imposed by the Fed in the
aftermath of the financial crisis. Testing showed the data to be useful. A regression of the next
3-month and 12-month returns revealed negative coefficients for both (i.e., returns were lower
following higher new equity financing), although the 3-month return was not statistically
significant (p = 0.112).
3 = + 1 33 +
12 = + 1 1212 +

15

The 12-month results were better, with the coefficient significant at the 5% level (p =
0.037), with a higher R-squared. See Appendix 2.

Secondary Distributions is a smaller series than New Equity Financing, but

even more current. It is weekly data with little lag. My source for Secondary Distributions is
Barrons weekly financial publication, which gets it from Dealogic LLC, New York City23.
Secondary distributions are often used by major stockholders to eliminate or reduce holdings,
i.e., investment liquidation. Chart 8 below shows a 52-week moving average of secondary
distributions, compared to the NYSE Composite Index. With one big anomaly, 2008, the chart
mirrors the market, with a very low reading just after the 2003 market low and very high
readings near market tops. The anomaly is 2008-2009, when, similar to the New Equity
Financing data, financial companies were forced to finance in a depressed market. Dealogic
evidently classified some of the bank financing as secondary offerings rather than primary
offerings. Aside from the 2008-2009 experience, the message is clear: investment activity (in
this case, corporations activity in their own stock) shows liquidation at high prices, usually
against a strong trend.

16

17

C. SPECULATIVE DEMAND
Margin debt is a measure of speculative demand. Traders, professionals (including hedge
funds) as well as the public, often execute their trades on margin, using the leverage to increase
their exposure. While a portion of margin debt does reflect short selling, the bulk of the
movement in margin debt is a function of traders expanding and contracting long positions.
Since traders are motivated by the trend (my premise), I expect to see this series mirror the
market, with low readings near market bottoms and high readings near market tops. Chart
9 below shows the market value of margin debt and margin debt as a percent of market value,
compared to a proxy for total market value, the Dow Jones U.S. Total Stock Market Index. 24
The total margin debt series and margin debt as a percent of market value both show a secular
rise. We expect to see the margin debt percent to rise more sharply late in a market up-cycle,
reflecting increased speculative demand.25 In the period from 1950s through the 1980s (not
shown in the chart) the margin debt percent tended to move between 1% and 2% most of the
time, troughing near the highs of bull markets and peaking close to the lows of bear markets,
the performance I would expect. Since the 1990s, the margin debt percent peaks were 1.96 in
February 2000 and 2.60 in July 2007 (the high so far in the advance from the March 2009 low
was in April 2013). We note that the low after the 2007 peak was a 1.94 in July 2012, just under
the 2009 low of 1.97, registered in July 2009.

18

D. INVESTMENT DEMAND
Insider activity represents investment demand (and supply). Insiders, who acquire the bulk
of their stock through new incorporations, rarely buy in the open market. As a result, there is
almost always more insider selling than insider buying, with an average of more than four times
as many sales as purchases in the last 10 years. But when they do buy in the open market, it is
usually at a low price. Insiders, by law, are longer term investors; if they take a short term
profit, they must return it. Chart 10 below is an 8-week moving average of the sell/buy ratio of
insider transactions, compared to the S&P 500 (red line).26 The bearish and bullish lines are 1
standard deviation above and below the 10-year average. With the 10-year average rising from
below 2.00 to above 4.00 since 2000, the bullish line has risen from under 1.00 to almost 2.00
and the bearish line has risen from under 3.00 to over 6.00. Hence, the indicator does not have
to fall below 1.00 (i.e., actual net insider buying on an 8-week basis) for bullish readings. Note
that the indicator is based on insider transactions, not shares or value. The reason is the
number of insiders is considered more important. That is, 10 insiders buying 10,000 shares each
is more significant than 1 insider buying 100,000 shares. All the low points are near medium
term or major lows. Bullish or near bullish readings were registered in the fall of 2001 (an
interim low in the 2000-2002 bear market), late 2002-early 2003 (a major low), late 2008-early
2009 (actual net buying at a major low), and the fall of 2011 (an interim low in the post-2009
bull market). There is one anomaly, with net buying at high prices in early 2000, a function of
the huge number of call options granted in the 1990s and the scramble to exercise them prior
to expiration. The bearish extremes are usually just prior to medium term or major tops.
Clearly, these investors tend to be on the right side during reversal periods.27
I tested the data by regressing future 8-week returns for the S&P 500 on two dummy
variables. The first dummy (BULL) is 1 if the 8-week reading is below the bullish threshold and 0
otherwise. The second dummy (BEAR) is 1 if the 8-week reading is above the bearish level and
0 otherwise. Hence, the formula is:
= + 1 8 + 2 8 +

The results showed average 8-week returns of 2.9% for BULL and -1.3% for BEAR. Both
coefficients are statistically significant at the 1% level, very good. BULL is more than twice as
effective as BEAR, conforming to my belief that insider net buying is rare and usually bullish
when it occurs, while insider selling is normal and less interesting. See Appendix 3.

19

20

E. POTENTIAL DEMAND
Potential Demand is buying power on the sidelines and funds in other financial assets,
notably bonds. Part One discussed the equity part of the equation. In this section I look at
cash and money market fund assets as a source of demand for stock. I have created an
Individual Buying Power Index from the Federal Reserve Flow of Funds statistics, using the
Households and Nonprofit Organizations (L.100) as a proxy for individuals. Individual
Buying Power is money market fund assets (Line 6) divided by money market fund assets
Line 6) plus equities (Line 18) plus mutual funds (Line 19).19 Bonds are not included in the
formula under the assumption that a shift in asset allocations (See Part One, Section 2
above) is a different decision than putting ready cash to work. Logically one would expect
buying power to be highest near major market bottom areas and lowest near major market
top areas and that appears to be the case. The secular rise in the buying power index is a
function of the secular rise in money market fund assets as public investment funds were
moved from stocks and bonds. Nevertheless, Chart 11 below shows rising cash into the
market bottoms in 1982, 2002-2003, and 2008-2009, and cash falling into the 1999-2000
and 2007 tops. The 15.84% reading in March 2009 was a record, well above the 13.00%
reading of September 2002, near the beginning of the 2002-2003 market bottom. After
declining steadily (and expectably) since the 2008-2009 market bottom, the latest available
reading of 5.34% in mid 2013 is the lowest since mid 1984. On the margin, there appears to
be public investment selling at low prices and public investment buying at high
prices, contrary to my belief that investors are usually on the right side near market turning
points, but that activity is probably dominated by professionally-managed foundations and
some public trading. The vast bulk of the money, 85% to 95%, remains in stock. Note that
the biggest shifts in the buying power index occur in the months right before and after
major market lows when traders (i.e., short-term-oriented market participants) panic out
and rush back into stocks.

21

22

F. POTENTIAL AND ACTUAL DEMAND THROUGH MUTUAL FUNDS

Charts 12 to 18 are derived from the monthly Research and Statistics report from the
Investment Company Institute (www.ici.org).28 The data show cash and cash flows for
equity mutual funds. The bulk of the buying and selling in equity mutual funds is
investment (i.e., long term) activity, but just as I noted for the Individual Buying Power
above, there does appear to be some trading (i.e., short term) activity. In other words,
while the bulk of equity mutual fund activity is conducted for long term objectives (demand
for IRAs and 401ks, for example), a small portion is trading activity, usually by traders
liquidating stock during bear markets and replacing stock during bull markets. Chart 12
below is cash as a percent of net assets, compared to the S&P 500. The chart shows rises
and falls during bear and bull markets, as one would expect. The cash percentage generally
swung from 5%-6% at market tops to 12%-13% at market bottoms in the 1970, and 1980s,
with earlier data in the 1950s similar.29 There is one big anomaly. Cash fell from 12.93% in
October 1990 to 4.04% in March 2000, but since then at market bottoms the figure has
been no higher than 5.90% (in February 2009, during the 2008-2009 market bottom). Fund
managers have been directed by superiors to stay much more fully invested and investors
have to make their own decisions to raise cash or not, by switching from equity mutual
funds to money market funds. Nevertheless, the data still show high cash at market
bottoms and low cash at market tops, in a new range of 5% to 6% at market bottoms
and 3% to 3% at market tops. It troughed at 3.5% during the mid 2007 market top and has
been as low as 3.3% in 2013.

23

Cash on hand is important, but cash flows are even more important. Most of the
time cash flows into and out of equity mutual funds represent public investment activity.
Chart 13 below shows sales (including reinvested dividends) and redemptions for equity
mutual funds, compared to the S&P 500, and Chart 14 below shows net sales (sales
including reinvested dividends minus redemptions plus net issuance of exchange-traded
funds, ETFs), compared to the S&P 500. Both sales and redemptions were in a secular rise
in the 1990s (and earlier data shows a secular rise in the 1950s-1980s as well). All three
series, sales (including reinvested dividends), redemptions, and net sales (including
reinvested dividends plus net issuance of ETFs) have flattened out since the 2000 market
top. Importantly, since the 2007 market top the net sales has barely been positive,
erratically swing from positive to negative. Chart 14 also belies the notion that ETF demand
has offset the sluggish demand for traditional equity mutual funds. ETF demand (i.e., the
net issuance of ETF shares), which I contend is very much the bailiwick of professional
traders, has only partially offset lack-luster equity mutual fund demand. In 1994-2007, net
flows into equity mutual funds, including ETFs, averaged $16.5 billion per month. In 20082012 (and 2013 through September), those flows averaged just $6.5 billion. Net public
investment demand (through mutual funds and ETFs) in the post-2003 bull market is far
below previous cycles, less than 40%.

24

Chart 15 below shows net exchanges. Net Exchanges is the flow from fixed
income funds (bond funds and money market funds) to equity funds, a positive flow, or the
flow from equity funds to fixed income funds, a negative flow. Typically, during advances,
especially late in advances, positive flows are predominant, while during declines, especially
late in declines, negative flows are predominant. In the up-cycles in the1990s flows were
positive most of the time, but the post-2003 up-cycle does not show persistent flows into
equities, a similar pattern to net sales discussed above. However, there were four
consecutive months of positive flows in 2013 (May through August), suggesting some trendfollowing demand.

25

Chart 16 below is total flows into equity mutual funds: sales including reinvested
dividends minus redemptions plus ETF net issuance (starting in January 2008; prior ETF data
is not significant) plus net exchanges. The picture shows persistent inflows in the 1990s upcycles and in the 2003-2007 up-cycle, but only erratic inflows in the post-2009 up-cycle.
The implication is that equity mutual fund managers had sizable impact on the demand for
stock in the 1990s and during the 2003-2007 up-cycle, but little impact since. Chart 17, the
net purchases of common stock by equity fund managers, below, shows just that. Hence,
insofar as most activity in equity mutual funds represents public investment activity, such
demand has been minimal in recent years. The rest of the equity mutual fund data reflect
public trading activity late in up and down cycles and professional trading in ETFs.

26

27

G. VALUATIONS
The last indicator in the long term (i.e., cyclical) category is valuations. Traditionally,
valuations have been the purview of fundamental analysis. Certainly, for the equity market
price/earnings ratios (P/Es) have been and remain a critical input for fundamental analysts
and strategists, and to a lesser degree, dividend yields are important. For the market as a
whole, fundamentalists believe that a measure of fair value in terms of P/Es can be
approximated by analyzing earnings growth rates and the level of interest rates.
Fundamentalists would argue that valuations reflect investment attitudes. The market is
worth a particular P/E multiple for each combination of growth rates and interest rates.
The so-called Greenspan Model is one such measure, comparing the earnings yield of the
S&P 500 to the rate on the 10-year treasury note.30 There are other models that include
inflation in the formula. Nevertheless, there are valuation levels that cannot be explained
by those fundamental investment formulas. During periods when traders are extremely
bearish or extremely bullish, they overwhelm investment attitudes and behavior.
Chart 19 below shows price/earnings multiples and yields on the S&P 500, compared to
the S&P 500 index, from 1976 to 2013. Note that in the 1990s the P/E multiple rose to the
mid 20s in the middle of the decade and then soared to unprecedented levels above 40 at
the end of the decade. Historically, there were periods when interest rates were
comparable or lower and earnings growth rates were higher, yet P/E multiples were far
lower. The 1950s is one example. A clear message is that there is an important
psychological component to valuations. In the early 1980s the S&P 500 P/E multiple was
below 10 and its yield was over 6%; in 1999-2000 the P/E multiple was in the 30-40 range
and its yield fell to barely over 1%. Why? Despite investment formulas that said stocks
were very cheap or very expensive, short term participants, motivated by very weak or
very strong trends, sold cheap stocks or bought expensive ones. Hence, there are
periods when cheap stocks stay cheap and periods when expensive stocks get more
expensive. Knowing who the traders are and what they are doing, and knowing who the
investors are and what they are doing trumps the formulas.

28

29

PART THREE
MEDIUM TERM INDICATORS
A. OPTION VOLUME INDICATORS
By definition, participants in the options markets are making short term judgments,
trading judgments. Even though they may be using options to augment a longer term
strategy, the limited life of an options contract (most of the activity in the options market is
in contracts that are just two or three months from expiration) means most participants are
making some kind of a short term decision when they buy or sell a call or a put. I recognize
that option activity includes hedging trades and not just directional trades, so the
movement of put/call ratios is not just a function of market psychology. Nevertheless, on
the margin, I believe shifts in put/call ratios should have a distinct sentiment component. If
option traders are becoming bearish, they would be expected to be more interested in puts
than calls. If option traders are becoming bullish, they would be expected to be more
interested in calls than puts. That appears to be the case. Put volume increases relative to
call volume when trend-motivated option traders grow more bearish and put volume
decreases relative call volume when option traders grow more bullish. Under my
assumption that short-term-oriented equity market participants are likely to be overly
bearish near lows of consequence and overly bullish near highs of consequence, put/call
volume ratios should often be found high near bottoms and low near tops.31 In other
words, puts garner more attention too late in declines and calls garner more attention
too late in advances. Remember, this is not to say options participants cannot be
successful; as a group they can be successful most of the time. However, their trendmotivated activities results in too much bullishness near tops and too much bearishness
near bottoms. Much of the activity in the options markets is professional; professionals
succeed by getting back on the right side quickly.
The four put/call option volume charts below have necessarily limited history since they
are daily charts; they start at the beginning of 2011 and end in the third quarter of 2013.
Earlier data is similar. I have drawn bull lines one standard deviation above one-year
averages and bear lines one standard deviation below one-year averages to give some
approximation of high and low levels, not because any particular level is a clear buy or sell
signal. The indicators should be used to increase or decrease confidence in the trend
outlook, not to anticipate imminent reversals. The charts are 10-day moving averages,
plotted along with the S&P 500. The four charts are equity options (opening transactions
only) on the ISE exchange (www.ise.com/market-data/isee-index), index options (opening
transactions only) on the ISE exchange, equity options (all transactions) on the CBOE
30

exchange (www.cboe.com/data/mktstat.aspx), and index options (all transactions) on the

CBOE. I have excluded the trading in VIX options since it can be argued that buying VIX calls
is a bearish bet, not a bullish bet. Note that even though the ISE data are just open-buyputs divided by open-buy-calls and the CBOE data also include sell-to-open calls and sell-toopen puts, the ratios are similar. The charts show that put/call ratios usually spike near
bottoms, bullish readings, implying the option participants are very negative. The ratios are
usually in the neutral-to-low regions during advances (i.e., the option participants are
getting the uptrend message correctly). The combination of low ratios and faltering price
action are bearish readings. Chart 19 below, the ISE 10-day Index Put/Call Ratio, gives the
least useful information, although the mid September 2012 very low reading occurred at
the inception of a two-month market correction. The very high readings in mid July 2011
and in the first half of July in 2013 were not helpful.

31

Chart 20 below, the ISE 10-Day Equity Put/Call ratio, was quite helpful, with all the very
high readings occurring near important or interim market bottoms. The sustained high
readings from mid August 2011 through late November 2011 coincided with the
intermediate bottom in the market in 2011. The sustained very high readings between mid
May 2012 and mid June 2012, and between late June 2013 and early July 2013 coincided
with interim market setbacks.

Chart 21 below, the CBOE 10-Day Index Put/Call Ratio, shows two clear, sustained high
readings, late August 2011, at the beginning of the market basing pattern that year, and in the
second half of May 2012, at the end of an interim correction. Once again, I find low index
put/call ratios near interim tops, such as mid March 2012 and mid May 2013, but also in the
second week of August 2012 (the middle of an upleg) and in mid December 2011 (early in an
upleg).

32

33

Chart 22 below, the CBOE 10-Day Equity Put/Call Ratio, shows sustained high readings
three times, mid June 2011 (an interim low, albeit close to a medium term market top), mid
August 2011, at the beginning of a medium term market base, and the third week of May
2012, at an interim market bottom. There are low readings in late March 2012, in the third
week of September 2012, an in late May 2013, all near the beginning of interim market
tops. There are a series of low or relatively low readings between mid October and mid
December 2013, all near minor tops, but no high readings near the interim minor bottoms
in that period.

I tested the put/call ratios, regressing 10-day returns for the S&P 500 on the 10-day
averages for each of the four ratios, the using the Newey West (1987) procedure. The equation
is:
= + 1 10 +

As I expected the equity ratios showed much better results than the index ratios (which
I attribute to much hedging activity in the index options and little hedging in the equity
options). For both the ISE and the CBOE data, high equity put/call ratios were associated with
34

higher next 10-day returns. For the CBOE equity put/call ratio, the coefficient is positive (0.045)
and significant (p = 0.025), and the R-squared is higher than the index put/call ratio (0.0135,
compared to 0.003). The results were similar for the ISE ratios.
A separation of the data into four quartiles by levels of the put/call ratios also showed
useful results. Low put/call ratios led to low returns and high put/call ratios led to high returns.
See appendix 4.

35

B. OPTION PREMIUM INDICATORS

Option premium indicators are analogous to option volume indicators. I noted above
that if options traders are bearish, put volume is expected to increase relative to call
volume, and if option traders are bullish, call volume is expected to increase relative to put
volume, and the data show both do occur. Likewise, traders should be expected to pay
relatively more for puts than calls when they grow bearish and they should be expected to
pay relatively less for puts than calls when they grow bullish. Put/call premium ratios
should move similarly to put/call volume ratios, and they do. For those comparisons, I use
the weekly option premium data from the Options Clearing Corporation (OCC)
(www.optionsclearing.com/webapps/weekly-volume-reports). The OCC provides a put/call
premium ratio for the equity options and a put/call premium ratio for the index options.
Charts 23 and 24 below show 4-week averages of the put/call premium ratios compared to
the S&P 500. I have drawn bull lines one standard deviation above the 10-year average of
the put/call ratios and bear lines one standard deviation below the 10-year averages. Chart
23 below, the equity premium ratio, shows several spikes in the 4-week average during the
2000-2002 equity bear market near interim lows. Very high readings were recorded several
times during the late 2002-early 2003 basing period and again at the beginning of the late
2008-early 2009 basing period. Readings were expectably low during the steady market
advance from the 2009 lows through 2013, although there was one increase to very high
readings in late June-early July 2013, associated with the end of a temporary market
correction. Chart 24 below, the index premium ratio, is similar, but gives an even clearer
message. It shows spikes to high levels near interim lows during the 2000-2002 equity bear
market (and one at a temporary low during the 1999 topping process) and near the lows of
temporary setbacks in the post 2009 bull market. The highest readings, the very high
sustained ratios were registered during the major bases in late 2002-early 2003 and late
2008-early 2009.

36

37

C. SENTIMENT POLLS

All of the indicators I have discussed above are transactional indicators, indicators that
measure what investors and traders are doing in the market place. There is another class of
sentiment indicators, surveys of equity participants, polls that measure what those
participants are saying about the equity market. As it turns out, all the polls are surveys of
equity market participants with short term time horizons. Based on my assumption that
short-term-oriented equity market participants, traders, are likely to be wrong at turning
points, I would expect the polls to show excessive pessimism at market bottoms and
excessive optimism at tops. And they usually do. I will discuss four polls that are widely
followed in the financial media: the poll of stock index futures traders conducted by
Consensus, Inc., the poll of stock index futures traders conducted by Market Vane, the poll
of market letter writers conducted by Investors Intelligence, and the poll of investment club
participants conducted by the American Association of Individual Investors.32
Consensus, Inc. and Market Vane, both polls of futures traders, have a slightly
different methodology so the readings are usually slightly different. Both surveys are polls
of professionals, but Market Vane weights the readings by its opinion of the impact of the
pollee, while Consensus, Inc. gives equal weight. Notably, both surveys are polls of
professional traders, people who make their living trading stock index futures; nevertheless,
they both show excessive optimism at market peaks and excessive pessimism at market
troughs. Each week the services poll the traders and report the percentage of bulls. In the
last 10 years Consensus, Inc. reports an average bull reading of 55.2%, with one standard
deviation of 15.5%, and Market Vane reports an average bull reading of 58.4%, with one
standard deviation of 9.9%. The Consensus, Inc. data in Chart 25 below shows a four-week
M.A. of the bullish percentage, with a bullish line drawn in one standard deviation below
the 10-year average and a bearish line drawn in one standard deviation above the 10-year
average, compared to the S&P 500. As I would expect, weak markets (and, therefore,
market lowpoints) are characterized by little optimism and strong markets (and, therefore,
market highpoints) are characterized by great optimism. After the major stock market top
in 2007, the indicator reached extremely low (i.e., bullish) levels as early as the end of
January 2008 and remained at or near such levels throughout the 2007-2009 bear market.
The absolute lowest reading of 20.3% was registered a week after the final market low was
reached in March 2009. Expectably, the indicator has been neutral most of the time during
the post 2009 bull market, with all but one of the excessively optimistic readings leading
to consolidations or interim setbacks, including October-November 2009, April-May 2010,
January-May 2011, mid January-mid May 2012, and September-October 2012. The
38

exception was mid February-mid June 2013, during which the market zigzagged higher.
Interestingly, the only fully bullish (i.e., excessively pessimistic) reading during the post
2009 bull market was after the biggest correction in that bull market, late August-late
October 2011.

39

As noted above, the Market Vane readings have been far less volatile in recent
years, although they have a similar pattern to Consensus, Inc. Chart 26 below shows that
the Market Vane bullish percentage reached its most negative levels during the 2006-2007
major equity market top, remaining at excessively optimistic levels from October 2006
through June 2007. It did not reach bullish (i.e., excessively pessimistic) during the 20072009 bear market, but its lowest reading was 34% right at the bottom in March 2009, very
near the bullish band, which stood at 33% at the market bottom. This indicator has been
neutral all the time during the post 2009 bull market, with a brief bearish exception, 1%
above the bearish band in the spring of 2013.

40

The oldest of the popular sentiment polls is the survey of market letter writers
conducted by Investors Intelligence. For over 50 years Investors Intelligence has been
reviewing market letters weekly and assigning a bull or bear or neutral opinion rating to
each comment and a giving aggregate figures. I have constructed a four-week average of
the ratio of bulls to bears; it is shown below in Chart 27, compared to the S&P 500. The
bearish line is drawn one standard deviation above the 10-year average bull/bear ratio and
the bullish line is drawn one standard deviation below the 10-year average. I call readings
outside the bands extreme or excessive. This indicator is popular because of its long,
readily available record (it was carried in Barrons for most of its history) and because it has
often been a contrary indicator. For example, Chart 27 below shows extreme pessimism
(i.e., a bullish reading) during the market basing patterns in late 2002-early 2003 and late
2008-early 2009. Most letter writers claim to be giving long term advice, but clearly they
are very influenced by short term moves in the market. The chart shows that as the market
started up from those two bases, letter writers quickly became more optimistic, typical of
short term traders. The charts do show extreme optimism at the 2000 and 2007 tops, but
they also show periods of great optimism prior to interim setbacks during the 2003-2007
and post 2009 bull markets. Hence, no formula is going to provide a clearly profitable
trading rule. Nevertheless, I believe this indicator can be used like other sentiment
indicators, to increase or decrease confidence in a trend interpretation. In other words,
when excessive optimism appears after an advance, look for a setback. If the ratio drops
quickly back to levels of extreme pessimism, it probably means the setback was a minor,
temporary affair. Such was the case in the summer of 2010 and in the fall of 2011. The
opposite is also true. When excessive pessimism appears after a decline, look for a
rebound. If the ratio rises quickly back to levels of extreme optimism, it probably means the
rally is a minor, temporary affair. Such was the case in July 2001, in January 2002, and in
April-May 2002.

41

42

The final poll is the survey of investment club participants conducted by the
American Association of Individual Investors (AAII). The AAII surveys investment clubs
weekly and tallies the number of bulls, bears, and neutral. The poll is flawed compared to
the other three polls above, because the list of pollees is not the same every week.
Nevertheless, since the results are similar to the other polls and because of the accessibility
of the data (it appears in Barrons every week), I include it in this paper. Chart 28 below
shows a four-week moving average of the bull/bear ratio compared to the S&P 500. The
bearish line is one standard deviation above the 10-year average and the bullish line is one
standard deviation below the 10-year average. The results are comparable to the other
polls with one exception. It is true that the most sustained negative readings (i.e., excessive
optimism) were recorded around the 2000 major stock market top and sustained positive
readings (i.e., excessive pessimism) were recorded at the major bottoms in late 2002-early
2003 and late 2008-early 2009, and the indicator did not reach levels of extreme pessimism
(falsely) during the 2000-2002 bear market. However, this indicator did not give a negative
signal (i.e., did not register extreme optimism) at the 2007 major equity market top.
To test the series, I organized the 1987-2014 weekly data by deciles for both the percent
of bulls and the percent of bears. The top decile of bulls and the bottom decile of bears
were considered excessive optimism. The bottom decile of bulls and the top decile of
bears were considered excessive pessimism. I tested short term periods (1, 2, 4, and 8
weeks) and, with some overlap, intermediate term periods (1, 2, 3, and 6 months). The
results were favorable for both bullish and bearish signals in the intermediate term horizon
and favorable for the bullish signals in the short term horizon. Only the bearish signals in
the short term horizon showed unfavorable results. The best results were for bullish
crossovers in the intermediate term horizon, with 80% of the signals registering positive
results, for a cumulative gain of 459% in the whole time period. If a few anomalous
readings in August-October 1987 are disincluded, the short term bearish signals would have
been much better. Those results were in line with my expectations: I believe bottoms form
quickly and tops are drawn out.

43

44

SUMMARY AND CONCLUSION

THE RELATIONSHIP BETWEEN SENTIMENT AND SUPPLY/DEMAND INDICATORS
AND OTHER TECHNICAL MEASURES
It was not my intent to produce a stock market trading model based solely on sentiment
and supply/demand indicators. They are, after all, supporting indicators to the basics of
technical analysis, trend and momentum indicators. I would never require all the sentiment
indicators to reach extremes at the same time to reach a conclusion, nor would I expect them
to do so. They are, after all, samples of attitudes and behavior, and any indicator could at times
be distorted and misleading. I believe there are four kinds of technical indicators. Trend and
momentum indicators, the most important technical indicators, illustrate the direction and
force of a move. But they often do not tell much about where the market is in a trend. The
second or third leg of advance, or the final leg of advance, may not look much different from
the first leg. Sentiment and supply/demand indicators, the second and third kinds of technical
indicators, give such insights and, accordingly, add to or subtract from confidence in a trend.
The sentiment and supply/demand indicators discussed above underscore the difference in
attitudes and behavior between investors and traders. The indicators show extreme investor
optimism and excessive trader pessimism at stock market bottoms and they show excessive
investor pessimism and extreme trader optimism at stock market tops. The best example of
that notion is to compare insider activity (see Chart 10 above) to option activity (see Charts 1924 above). Bottoms of consequence are all characterized by insider optimism, low insider
sell/buy ratios, and option trader pessimism, high put/call ratios. If traders stay cautious and
investors stay bullish in an advancing trend, I have more confidence in the durability of the
uptrend. If those indicators look more or less like they did at the inception of an uptrend
months after the uptrend began (I.e., bullish investors and reluctant traders), I continue to have
high confidence in that uptrend. If a setback results in a quick, sizable increase in trader
pessimism, I have high confidence the setback is likely to be a brief, temporary affair. If traders
are slow to get pessimistic in a downtrend after a stock market top is completed, I have high
confidence in the sustainability of the downtrend. If the indicators look more or less like they
did at the inception of a downtrend months after a downtrend began, I continue to have high
confidence in that trend remaining down. If a rebound results in a quick, sizable increase in
trader optimism, I have high confidence the rally is likely to be a brief, temporary affair.
There is a fourth kind of stock market technical indicator, with which I am not concerned
here: intermarket analysis, the relationship between equities and bonds (actually all fixed
income investments), commodities, and currencies. All other things being equal, the change in
price of an alternative to equities will change the supply/demand picture for stocks. For
example, if the price of bonds goes up, the supply/demand curve for stocks will rise (i.e., bonds
45

become relatively more expensive); if the price of gold goes down, the supply/demand curve
for stocks will fall (i.e., gold becomes relatively cheaper). A thorough discussion of intermarket
analysis can be found in John Murphys Intermarket Analysis, and additional comments are
available in John Murphys Technical Analysis of the Financial Markets, and Martin Prings
Technical Analysis Explained.33

46

FOOTNOTES, GLOSSARY, AND REFERENCES

g. See the Glossary below Footnotes.

1.
2.
3.
4.

Chan, Jegadeesh, and Lakonishok, Financial Analysts Journal, November, 80-90, 1999.
Prechter and Parker, Journal of Behavioral Finance, 84-108, 2007.
Lo and MacKinlay, May 1989. NBER Working Paper No. w2168.
3. Lo and Wang, The Review of Financial Studies Summer 2000. Vol. 13. No. 2, pp.257300 2000 The Society for Financial Studies.
5. 4. Osler, Carol L., Economic Policy Review, July 2000, Osler Carol L., Journal of Finance,
Vol. LXIII, No. 5 October 2003, and Mizrach and Weerts, Highs and Lows: A Behavioral
and Technical Analysis (November 27, 2007).
6. Papailias and Thomakos, September 2011.
7. Lo, Mamayski, and Wang March 2000, NBER Working Paper No. w7613, Osler and Chang
August 1995 FRB of New York Staff Report No. 4, Savin, Weller, and Zvingelis Journal of
Financial Econometrics Spring 2007, and Weller, Friesen, and Dunham University of
Nebraska-Lincoln August 2007.
8. Lachhwani and Khodiyar, Quest-Journal Of Management and Research August 2013.
9. Magazzino, Mele, and Prisco, Journal of Money, Investment, and Banking March 2012.
10. Lo, Maymaski, and Wang: Foundations of Technical Analysis: Computational Algorithms,
Statistical Inference, and Empirical Implementation. Journal of Finance v55 (4 Aug.) pp.
1705-1765. HS=Head-and-Shoulders, BBOT=Broadening Bottom, RTOP=Rectangle Top,
RBOT=Rectangle Bottom, and DTOP=Double Top. The other patterns tested in which
they found no significant results were IHS=Inverted Head-and Shoulders,
BTOP=Broadening Top, TTOP=Triangle Top, TBOT=Triangle Bottom, and DBOT=Couble
Bottom.10,
11. Lo, Andrew: Heretics of Finance, 2010.
12. A brief discussion of academias Behavioral Finance terms can be found in Thinking, Fast
and Slow by Daniel Kahneman, pages 119-128, 154, 284, 289-299, 349, 417-418, 427430, 444, 445, and 471-472. Farrar, Straus, and Giroux, 2011. Nobel Laureate
Kahneman is generally credited with being among the first to define anchoring, along
with Amos Tversky.
13. A fascinating look on the life of Jesse Livermore is Reminiscences of a Stock Operator by
Edwin Lefevre. Wiley 1997.

47

14. Mackay, Charles, Extraordinary Popular Delusions and the Madness of Crowds. pp. 2850 and 50-54.
15. Greenspan, Alan, Speech to American Enterprise Institute, 1996.
16. Shiller, Robert, Irrational Exuberance, 2000.
17. Paul Macrae Montgomery (Montgomery Capital Management) popularized the notion
of the magazine cover indicator. Paul Krugman was quoted on March 28, 2009: Whom
the Gods would destroy, they first put on the cover of Business Week.
18. Commentary on many of my key indicators outside the behavioral area, such as trend,
momentum, relative strength, and intermarket measures, can be found in an interview
with me in Technically Speaking, by Chris Wilkinson. Traders Press 1997.
19. Kirkpatrick & Dahlquist, Technical Analysis, pp. 181-182.
20. The data for charts 1, 2, 3, and 11 come from the Federal Reserve Statistical Release,
Z.1, Second Quarter 2013.
21. I credit investment strategist Michael Sherman, with whom I worked in the 1980s at
Shearson Lehman Hutton, for pointing out to me the utility of some of the Flow of Funds
statistics from the Federal Reserve. See also Technically Speaking, p. 324.
22. SIFMA U.S. Foreign Activity Report Second Quarter 2013.
23. Kirkpatrick & Dahlquist, Technical Analysis, p. 110, p. 180.
24. New York Stock Exchange, Fact Book, 2013.
25. Pring, Technical Analysis Explained, pp. 501-505; Kirkpatrick & Dahlquist, Technical
Analysis, pp. 110-111 and 179; and Wilkinson, interview with Philip J. Roth, Technically
Speaking, p. 332.
26. Vickers Stock Research, Weekly Insider Reports.
27. Pring, Technical Analysis Explained, pp. 492; Kirkpatrick & Dahlquist, Technical Analysis,
118; and Wilkinson, interview with Philip J. Roth, Technically Speaking, p. 336.
28. Investment Company Institute, Research & Statistics, Monthly Trends in Mutual Funds
Investing; and Monthly Exchange-Traded Fund Data.
29. Pring, Technical Analysis Explained, pp. 499-501; Kirkpatrick & Dahlquist, Technical
Analysis, pp. 108-109; and Wilkinson, interview with Philip J. Roth, Technically Speaking,
pp. 326 and 328.
30. Kirkpatrick & Dahlquist, Technical Analysis, p. 192, attributing its discovery to Ed Yardeni
(www.yardeni.com). Additional comments on valuation as a measure of psychology can
be found in Pring, Technical Analysis Explained, p. 522 and Wilkinson, Interview with
Philip J. Roth, Technically Speaking, pp. 324-325.
31. Kirkpatrick & Dahlquist, Technical Analysis, pp. 97-100; Pring, Technical Analysis
Explained, p. 506; Pring, Investment Psychology Explained, pp. 134-153; and Wilkinson,
interview with Philip J. Roth, Technically Speaking, pp. 333-334.

48

32. Consensus Inc., Independence, Missouri; Market Vane, Pasadena, California; Chartcraft,
Investors Intelligence, New Rochelle, New York; American Association of Individual
Investors, Chicago, Illinois.
33. Murphy, Intermarket Analysis; Pring, Technical Analysis Explained, pp. 559-661; Murphy,
Technical Analysis of the Financial Markets, pp. 413-431.

49

A Glossary of Terms Used in Trading and Technical Analysis

Accumulation

Process by which an excess supply of stock is absorbed by an expanding demand that over a period of
time has favorable effect on the price. Generally, it is a period of price equilibrium following a decline.

Advance-Decline Line

Number of stocks advancing divided by the number of stocks declining over a particular time period.

Advisory Services

Newsletters that comment upon and attempt to forecast the course of various markets. They tend to
be trend followers and are often overly bullish at tops and overly bearish at bottoms.

Apex

Point of intersection of two trendlines; the usual connotation that some new trend may evolve as
prices approach that intersection.

Arbitrage

Simultaneous buying in one market and selling in another market in order to take advantage of
differences in price.

Back and Fill

Same as Consolidation. See also Stabilization. Especially when using point and figure charts.

Bar Chart

Price/time chart that depicts high, low and close data as bars on the vertical axis and time intervals on
the horizontal axis. Volume is usually indicated also, as vertical bars on the bottom of the charts
under the applicable price data.

Base

Period of accumulation (see above); Also called Bottom.

Bear Market

A long period of time, often a year or more, when the general trend of securities prices is down.

Bear Trap

A false move to the downside that does not initiate a new downtrend, but is actually the final reaction
prior to a sustained advance, hence trapping the bears.

Block

A single transaction of large number of shares. The NYSE defines transactions of 10,000 shares or
more as blocks.

Blow Off

The final phase of an uptrend, ending a mark-up phase, when prices rise very rapidly usually on high
volume, leading to a sharp reaction. See Climax.

Bottom

Period of accumulation. See Base.

Breadth

Net number of stocks advancing versus declining. When advances exceed declines, breadth is
positive; when the reverse is true, breadth is negative.

Breakdown

Like Breakout, but used to refer to the downside.

Breakout

When stocks price or volume exceeds previously recorded high or low (or resistance or support level)
or some other predetermined criteria. Also called Penetration.

Bull Market

A long period of time, usually a year or more, when the general trend of securities prices is up.

Bull Trap

A false move to the upside that does not initiate a new uptrend, but is actually the final rally before a
sustained decline, hence trapping the bulls.

Churning

Hesitation in a trend, usually leading to a reaction. Volume is normally relatively high (after an
advance), with limited price progress.

Climax

(Climatic lowpoint or climactic highpoint). A sudden end to a trend, accompanied by high volatility
and high relative volume.

Concession or Spread

Price differential between trades when blocks of stock are involved.

Confirmation

Two or more trend or momentum measures extending their trends to new highs (or lows) at the same
time. The implication is confidence that the trend will continue. Often said of the Dow Jones
Industrial Average versus the Dow Jones Transportation Average.

Consolidation

A pause in a trend with the expectation the trend will be resumed in the same direction. Usually
50

considerably shorter in time than Stabilization.

Continuation Pattern

Consolidation phase that temporarily interrupts an up or down move and sets the stage for another
move in the same direction later on, usually of short duration.

Correction

A movement in the opposite direction of a trend that does not break or reverse that trend. Usually
longer in time than a reflex rally or reflex reaction.

Cyclical

Same Long Term. See below.

Discount

The amount by which the price of an option or future falls below the theoretical value, representing
the degree of pessimism of market participants.

Distribution

Process in which demand for a stock is overcome by an expanding supply, which over a period of time
has an unfavorable effect on the price of a stock. It is generally a period of trend neutrality in price,
but often has high volatility.

Divergence

One measure failing to extend its trend to a new high (or low) while another measure does. This leads
to less confidence in continuing the trend. Often said of the Dow Jones Industrial Average versus the
Dow Jones Transportation Average. Opposite of confirmation.

Downtick

A transaction that occurs at a lower price than the previous transaction.

Elliott Wave Theory

A trend assessing technique which emphasizes FORM over magnitudes, hence WAVE.

Extended

When a stock has advanced or declined to, or in excess of, its trend parameters and a consolidation is
anticipated.

Failsafe Point

See Stop Point.

Financial Futures

Futures contracts based on financial instruments such as U.S. Treasury Bonds, Fed Funds and other
interest rate-sensitive issues, currencies and stock market indices.

Futures Contract

Exchange-traded contracts that give the holder the obligation to buy or receive a certain amount of a
product at a specific prices on a specific date. Futures are used by business as a hedge against
unfavorable price changes and by speculators who hope to profit from such changes.

Gap

When a stocks high and low prices on a given day do not overlap the stocks high and low of the
previous day. When a gap initiates a trend, it is called a breakaway gap; an exhaustion gap ends or
reverses a trend, and measuring gaps usually duplicate the most recent move before the gap.

Insiders

Officers, directors, shareholders and others privy to non-public information that affects a companys
stock.

Liquidation

Phase following distribution, when prices decline relatively easily.

Liquidity

Ability of the market to absorb significant increases in volume with minor price fluctuations.

Long Term

More than six months; usually one complete bull-bear cycle, normally averaging four to five years, the
trend associated with a normal economic cycle.

Market Analysis or
Technical Analysis

The study of supply/demand relationships, investor psychology, monetary changes, and the study of
price and volume movements of the market as a whole, and individual stocks in particular, in order to
determine the probability of direction and degree of future price movements. Whereas fundamental
analysis is concerned with the company (and its sales, earnings, products, management, etc.),
technical analysis is concerned only with the stock of the company (i.e., the changes in the
supply/demand relationship for the stock in the market place).

Mark Up

Phase following accumulation when prices rise relatively easily.

51

Medium or
Intermediate Term

Five weeks to six months; usually three to five months. One leg of a bull or bear cycle.

Momentum Indicators

Market indicators generally related to price and volume, which attempt to determine overbought and
oversold conditions and the underlying strength or weakness of current market trends. Most are
moving averages.

Money Flow

Price multiplied by volume accumulated overtime, with the price change determining positive or
negative flow. Derived from Tick Volume (see below).

Moving Average

Average of price or volume over a period of time that is used to smooth trends of minor fluctuations.

Moving Average
Convergence-Divergence

A momentum technique that utilizes two exponential-weighted moving averages, looking for crossing
points between them and overbought-oversold conditions in them.

Near Term

(Also Short Term) Up to two months; usually three to five weeks.

Odd Lot

Units of stock less than 100 shares.

Odd Lot Shorts

Units of usually less than 100 shares that are sold short. Since these are usually traded by trendfollowing individuals (as opposed to institutions) who join a trend very late, a high ratio of odd lot
short sales to odd lot sales often signals a market bottom. Conversely, a low ratio of odd lot shorts
often signals an impending market top.

Odd-Lotter

Any uniformed trader (who may also trade in round lots).

On-Balance-Volume

A cumulative volume figure, the direction of which depends on price movement; ie, a positive change in
price indicates the volume for that particular time interval is positive; a negative price change indicates
negative volume.

Opening Block

The first transaction of the day in a particular stock, which usually represents an accumulation of
individual buy and sell orders but that appears on the tape as a single block.

Option

The right (but not the obligation) to buy or sell specific quantity of a specific security at a certain price
within a specific time. An option giving the buyer the right to purchase a security at a fixed price is a
call option; a option giving the buyer the right to sell the security at a fixed price is a put option.

Oscillator

Indictor that measures the strength of momentum (or sentiment) activity by moving to a positive or
negative extreme around zero. Moving averages are usually used for the calculations.

Overbought

Recently extended or exceeded trend parameters on the upside.

Oversold

Recently extended or exceeded trend parameters on the downside

Penetration

See Breakout.

Point and Figure

A method of recording price activity without reference to time that is concerned only with price
trends and changes in price trends.

Premium

The amount by which the price of an option or future exceeds the theoretical value, representing the
degree of optimism of market participants.

Price Potential or Objective Technical appraisal of a stocks future value. May be near-, medium- or long-term; an upper or lower
boundary of a trend channel; a point and figure projection; or the next significant support or
resistance level.
Primary Offering

A public sale of newly issued stock.

52

Program Trading or
Program Activity

Buying or selling packages of stocks. Often done by institutions making a strategy change or index
funds moving into or out of the market. When such trading in stocks is accompanied by offsetting
trades in index futures, index options or stock options, it is called arbitrage program activity.

Protection Point

See Stop Point.

Pullback

See Reflex Reaction.

Questionable

Deteriorating; said of a trading range when a support violation is anticipated.

Rate of Change

A momentum indicator which compares a current price to an historical price, often 14 days (although
any time period can be used) to determine strength or weakness, and overbought and oversold
conditions.

Rebuilding

See Stabilization.

Reflex Rally or
Reflex Reaction

Movement against a stocks trend that does not break or reverse the trend buy merely corrects an
overbought or oversold condition. Also called Pullback or Throwback.

Relative Strength Line

The line that represents a stocks (or groups) price performance compared with a broad market index
(or a group or another stock). It is computed by dividing the price of a stock by the market index for
each time period and plotting a line of the resulting ratios.

Resistance

A price level at which selling is expected to exceed demand and temporarily stop or reverse an
advance. Opposite of Support.

Reversal Pattern

A characteristic pattern that signals the end of an up or down move.

RSI

Not to be confused with Relative Strength Line (see above), it is momentum indicator which utilizes
the NUMBER of and down days in a particular time period to determine strength or weakness, and
overbought and oversold conditions.

Secondary Distribution

A public sale of an already existing block of stock, usually sold by an insider or other large holder.

Secular Trend

The trend that encompasses two or more cyclical trends (also two or more economic cycles) and often
lasts 10 to 20 years or more. Also called the Super Cycle.

Sentiment Indicators

Indicators that are used to try to gauge changes in the psychology underlying investment decisions.

Shakeout

A sharp reaction (usually in an uptrend or base) that temporarily violates a trendline or support level,
but quickly reverses to the original trend. Traders are shaken out of long positions on the violation
but stronger holders (investors) retain their positions.

Short Covering

The purchase of previously shorted stock.

Short Selling

Selling stock that is not owned (usually borrowed), normally in anticipation of a price decline.

Stabilization

A period of sideways price action, usually before a trend is reversed. Sometimes called basing or
accumulation after a decline, and top formation or a distribution after an advance. Also call
Rebuilding.

Stochastics

A momentum indicator which utilizes the closing price relative to the range in a particular time period
to determine strength or weakness, and overbought and oversold conditions.

Stop Point

The price level which, if exceeded (on the downside in a long position; on the upside in a short
position), breaks the trend or negates a reversal pattern. That point is usually below a support level
or just above a resistance level. Sometimes called Failsafe Point or Protection Point.
53

Support

A price level at which buying is expected to increase enough to temporarily stop or reverse a decline.
Opposite of Resistance.

Testing

Consolidation with the expectation that a support (or resistance) level will hold.

Theoretical Value

The worth of an option or future, calculated by considering (1) the difference between the options or
futures exercise price and the price of the underlying asset (stock, commodity, currency or index) and
(2) the time value of money. Also called Underlying Value.

Ticks

Net of upticks minus downticks.

Tick Volume

Similar to on-balance volume except that the base is not a time period but an individual transaction;
e.g., for any one day the net tick volume is computed by subtracting the total volume that occurred on
downticks from the total volume that occurred on upticks.

Top

Period of distribution.

Trading Index

The ratio of the advance/decline ratio to the upside/downside volume ratio. The index is favorable or
overbought when significantly less than 1.00, and unfavorable or oversold when significantly
morethan 1.00. Extremes are noted below 0.50 and greater than 2.00.

Trading Range

Upper and lower boundaries of a stocks or indexs price range. Normally, a neutral or horizontal
trend is implied.

Trend Channel

Two parallel trendlines that contain between them all or virtually all of the price movement data.

Trendline

Line that connects two or more points on a chart and represents the slope of movement.

Upside/Downside
Volume Ratio

Number of shares traded in advancing stocks divided by the number of shares traded in declining
stocks over a particular time period.

Uptick

A transaction that occurs at a higher price than previous transaction.

Volatility

Degree in intraday or interday fluctuation, not necessarily correlated with price movements over time.

Volume

Total trading activity in a stock or group of stocks for a specific time period, usually represented as
vertical bars at the bottom of a bar chart.

Wedge

A pattern of rising tops and rising bottoms in an uptrend or declining tops and declining bottoms in a
downtrend, where only marginal progress is made, usually leading to a trend reversal.

Zero Downtick

A transaction that occurs at the same price as the previous transaction, but lower than the last
different price.

Zero Uptick

A transaction that occurs at the same price as the previous transaction, but higher than the last
different price.

54

REFERENCES
AAII Index, American Association of Individual Investors, 625 N. Michigan Ave., Chicago, Illinois
60611.
Barrons, Market Laboratory, Weekly, 1978-2013.
Bauer, Richard J. and Dahlquist, Julie R., Technical Market Indicators: Analysis and Performance,
John Wiley & Sons 1999.
CBOE www.cboe.com/data.
Chan, Jegadeesh, and Lakonishok, The Profitability of Momentum Strategies, Financial Analysts
Journal, November 80-90, 1000.
Consensus, Inc., P.O. Box 520526, Independence, Missouri 64052.
Davis, Ned, The Triumph of Contrarian Investing, McGraw-Hill 2003.
Edwards, Robert D., Magee, John, and Bassetti, W.H.C., Technical Analysis of Stock Trends, CRC
Press 2013.
Elias, David, Dow Jones 40,000: Strategies for Profiting from the Greatest Bull Market in History,
McGraw-Hill 1999.
Federal Reserve Statistical Release, Z.1, Financial Accounts of the United States, Flow of Funds,
Balance Sheets, and Integrated Macroeconomic Accounts. Second Quarter 2013.
www.federalreserve.gov.
Frost, John A. and Prechter, Robert R., Elliott Wave Principle: Key to Market
Behavior, Elliott Wave International 1998.
Greenspan, Alan: Speech entitled The Challenge of Central Banking in a Democratic Society at
the Francis Boyer Lecture of the American Enterprise Institute for Public Research Policy
December 5, 1996.
Hoffmann, Arvid O.I.; Shefrin, Hersh; and Pennings, Joost M.E.; Behavior of Portfolio Analysis of
Individual Investors (June 24, 2010).
Investors Intelligence, Chartcraft Inc., 30 Church St., New Rochelle, New York 10801.
ISE www.ise.com/market-data.

55

Kirkpatrick II, Charles D., and Dahlquist, Julie R., Technical Analysis: The Complete Resource for
Financial Market Technicians, Second Edition, FT Press 2010.
Lachhwani, Hitendra and Khodiyar, Bhavesh Vishanji, Profitability of Technical Analysis: A Study
on S&P CNX Nifty (August 15, 2013) Quest-Journal of Management and Research, 3(2), 31-41.
Lee, Cheng-Few; Shih, Wei-Kang; and Chen, Hong-Yi; Technical. Fundamental, and Combined
Information for Separating Winners from Losers (April 15, 2010).
Lo, Andrew and Hasanhodzic, Jasmina, Heretics of Finance: Conversations with Leading
Practitioners of Technical Analysis, John Wiley & Sons May 2010.
Lo, Andrew and Hasanhodzic, Jasmina, Financial Prediction from Babylonian Tablets to
Bloomberg Terminals, John Wiley & Sons September 2010.
Lo, Andrew W. and Mackinlay, A. Craig, A Non-Random Walk Down Wall Street, Princeton, New
Jersey, Princeton University Press.
Lo, Andrew W., The Adaptive Markets Hypothesis: Market Efficiency from an Evolutionary
Perspective, Journal of Portfolio Management. August 2004.
Lo, Andrew W. and Wang, Jiang, Trading Volume: Definitions, Data Analysis, and Implications of
Portfolio Theory. The Review of Financial Studies Summer 2000 Vol. 13 No. 2, pp. 257-300.
2000 The Society for Financial Studies.
Lo, Andrew W., Mamaysky, Harry, and Wang, Jiang, Foundations of Technical Analysis:
Computational Algorithms, Statistical Inference, and Empirical Implementation (March 2000).
NBER Working Paper No. w7613.
Lo, Andrew W. and Mackinlay, A. Craig, Stock Market Prices Do Not Follow Random Walks:
Evidence from a Simple Specification Test (May 1989). NBER Working Paper No. w2168.
Mackay, Charles, Extraordinary Popular Delusions and the Madness of Crowds, London 1841.
Magazzino, Cosimo; Mele, Marco; and Prisco, Giancarlo; The Elliott Wave Theory: Is It True
During the Financial Crisis? March 2012 Journal of Money, Investment and Banking, 24, 100108.
Market Vane, P.O. Box 90490, Pasadena, California 91109.
Mizrach, Bruce and Weerts, Susan, Highs and Lows: A Behavioral and Technical Analysis
(November 27, 2007).

56

Montier, James, Behavioral Investing, A practitioners guide to applying behavioural finance,

John Wiley & Sons, 2007.
Murphy, John J., Intermarket Analysis, Profiting from Global Market Relationships, John Wiley &
Sons, 2011.
Murphy, John J., Technical Analysis of the Financial Markets: A Comprehensive Guide to Trading
Methods and Applications, New York Institute of Finance 1999.
NYSE www.nyse.nyx.com.
Options Clearing Corp. www.optionsclearing.com.
Osler, Carol L. and Chang, P.H. Kevin, Head and Shoulders: Not Just a Flaky Pattern (August
1995). FRB of New York Staff Report No. 4.
Osler, Carol L., Support for Resistance: Technical Analysis and Intraday Exchange Rates.
Economic Policy Review, Vol. 6, No. 2, July 2000.
Osler, Carol L., Currency Orders and Exchange-Rate Dynamics: Explaining the Success of
Technical Analysis (April 2001). FRB of New York Staff Report No. 125.
Osler, Carol L., Currency Orders and Exchange Rate Dynamics, An Explanation of the Predictive
Success of Technical Analysis, Journal of Finance, Vol. LXIII, No. 5 2003.
Papailias, Fotis and Thomakos, Dimitrios D., An Improved Moving Average Technical Trading
Rule (September 2011).
Prechter, Robert R. Jr., The Wave Principle of Human Social Behavior and the New Science of
Socionomics, Gainesville, GA: New Classics Library, 1999.
Prechter, Robert R. Jr. and Parker, Wayne D., The Financial/Economic Dichotomy in Social
Behavioral Dynamics: The Socionomic Perspective, Journal of Behavioral Finance 2007, Vol. 8,
No. 2, pp. 84-108.
Pring, Martin J., Investment Psychology Explained: Classic Strategies to Beat the Markets, Wiley
November 1995.
Pring, Martin J., Technical Analysis Explained, Fourth Edition, McGraw-Hill, 2002.
Savin, N. Eugene; Weller, Paul A.; and Zvingelis, Janis; The Predictive Power Of Head-andShoulders Price Patterns in the U.S. Stock Market (Spring 2007). Journal of Financial
Econometrics, Vol. 5, Issue 2, pp. 243-265.

57

Schiller, Robert J.: Irrational Exuberance, Second Edition, Princeton University Press, 2005.
Securities Industry and Financial Markets Association (SIFMA) U.S. Foreign Activity Report
Quarterly, through the second quarter of 2013.
Source Media, New York, NY www.sourcemedia.com
Vickers Stock Research, a subsidiary of Argus Research Group (www.vickers-stock.com)
Weller, Paul A.; Friesen, Geoffrey C.; and Dunham, Lee M.; Price Trends and Patterns in
Technical Analysis: A Theoretical and Empirical Examination (August 2007). University of
Nebraska-Lincoln Finance Department Faculty Publications.
Yardeni, Edward; Abbott, Joe; Johnson, Debbie; and Qunitana, Mali: Stock Market Indicators:
Fundamental, Sentiment, & Technical, Yardeni Research Inc., October 31, 2013.

58

APPENDICES
Appendix 1.
1. FLOW OF FUNDS: QUARTERLY DATA 1/1/1980 12/31/2013

59

Appendix 2.
NEW EQUITY FINANCING
Regression with
Newey-West standard errors
maximum lag : 5

Newey-West
Coef.
Std. Err.

ret3m
mtot3
_cons

-0.0004
0.0332

Number of obs =
450
F( 1, 448) =
2.53
Prob > F
=
0.1123

P>t

0.0003 -1.5900
0.0083 4.0200

0.1120
0.0000

R-squared = 0.0130

Regression with Newey-West standard errors

maximum lag : 5

ret12m
mtot12
_cons

Number of obs =
432
F( 1, 430) =
4.37
Prob > F
=
0.0372

Newey-West
Coef.
Std. Err.
-0.0005
0.1450

P>t

0.0002 -2.0900
0.0251 5.7700

0.0370
0.0000

R-squared = 0.0512

60

Appendix 3.
INSIDER ACTIVITY
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.161
R Square
0.026
Adjusted R Square
0.024
Standard Error
0.060
Observations
1233

Intercept
BULL DUMMY
BEAR DUMMY

Coeff. Std. Err. t Stat P-value

0.013
0.002
6.745
0.000
0.029
0.007
4.394
0.000
-0.013
0.004 -3.029
0.003

Appendix 4.
PUT/CALL RATIOS
CBOE Data
Regression with
maximum lag: 5

spy10ret
indexpctm10
_cons

Newey-West

standard errors

Number of obs =
F( 1, 1053) =
Prob > F
=

1055
0.41
0.5215

Coef.

Newey-West
Std. Err.

P>t

-0.006
0.016

0.010
0.015

-0.640
1.050

0.521
0.292

Newey-West

standard errors

Number of obs =
F( 1, 1053) =
Prob > F
=

1055
5.02
0.0252

R-squared = 0.003

Regression with
maximum lag: 5

61

Coef.

Newey-West
Std. Err.

P>t

0.045
-0.0224568

0.020
0.0124089

2.240
-1.81

0.025
0.071

spy10ret
equitypctm10
_cons

R-squared = 0.0135

ISE Data
Regression with
maximum lag: 5

Newey-West

standard errors

Coef.

Newey-West
Std. Err.

P>t

-0.009
0.022

0.005
0.008

-1.910
2.690

0.056
0.007

Newey-West

standard errors

Coef.

Newey-West
Std. Err.

P>t

0.045
-0.021

0.017
0.011

2.590
-1.970

0.010
0.050

spy10ret
iseidxtm10
_cons

Number of obs =
F( 1, 1053) =
Prob > F
=

1055
3.65
0.0689

R-squared = .0085
Regression with
maximum lag: 5

spy10ret
iseeqtm10
_cons

Number of obs =
F( 1, 1053) =
Prob > F
=

R-squared = .0018

10-day S&P 500 return quartiles

Index Data
CBOE
ISE

Mean 1
-0.032
0.022

Mean 4
0.045
0.008

Contrast
0.077
0.014

Std. Err.
0.022
0.022

t
3.50
0.65

P>t
0.00
0.51

Equity Data
CBOE
ISE

Mean 1
-0.062
-0.053

Mean 4
0.120
0.064

Contrast
0.183
0.117

Std. Err.
0.021
0.022

t
8.57
5.37

P>t
0.00
0.00

62

1055
6.73
0.0096

Appendix 5
AAII
The top decile for bears and the bottom decile for bulls were considered excessive pessimism
and the bottom decile for bears and the top decile for bulls were considered excessive
optimism. Two types of tests were conducted: one, the signal is generated when the series
moves into the extreme decile and two, the signal is generated when the series moves into and
then back out of the extreme decile.
Short Term Results
Bullish %: Top Decile Test (For Bearish Signals)
Bearish Cross

Date

Bearish - Cross Back

% Neg

47%

56%

49%

44%

% Neg

42%

48%

48%

43%

% Pos
Net
Ret

53%

44%

51%

56%

58%

52%

52%

57%

1%

0%

-16%

-46%

% Pos
Net
Ret

7%

-16%

-31%

-19%

spx

1W

2W

4W

8W

spx

1W

2W

4W

8W

3.99%

-1.95%

8/14/1987

333.99

-5.72%

1.57%
33.06%
21.98%

10/2/1987

328.07

0.57%
5.18%

-2.08%
13.83%

-3.27%
23.25%

-7.37%
29.80%

Date

8/7/1987

323

8/21/1987

335.9

3.40%
2.64%

9/18/1987

314.86

1.68%

4.20%

-7.55%
10.21%

8/18/1989

346.03

1.30%

2.23%

0.20%

-0.92%

8/11/1989

344.74

0.37%

1.68%

0.85%

4.37%

3/1/1991

370.47

1.21%

0.84%

0.22%

0.86%

2/22/1991

365.65

2.54%

0.50%

4.18%

3/22/1991

367.48

1.04%

3.04%

3.67%

1.31%

3/15/1991

373.59

1.32%
1.64%

-0.61%

2.03%

0.85%

4/12/1991

380.4

-0.36%

-1.23%

-0.48%

4/5/1991

375.36

1.34%

2.36%

1.45%

3.38%

5/3/1991

380.8

1.00%
1.33%

-2.21%

1.91%

-0.76%

4/26/1991

379.02

-0.87%

-0.40%

-2.13%

1/8/1993

429.05

1.89%

1.65%

4.63%

5.98%

12/18/1992

441.28

-1.57%

-1.39%

-1.81%

1/22/1993

436.11

0.61%

2.94%

-0.20%

2.93%

1/15/1993

437.15

0.37%

1.70%

3.27%

3/3/1995

485.42

3.15%

5.94%

488.11

0.30%

2.63%

5.08%

1/5/1996

616.71

-0.79%

3.10%

5.53%

12/29/1995

615.93

0.41%

-2.62%

1.35%

5.08%

2/9/1996

656.37

-0.90%

-2.49%

-2.16%

1/19/1996

611.83

1.60%

3.92%

5.91%

6.67%

5/24/1996

678.51

-0.94%

-1.42%

-7.61%

5/17/1996

668.91

1.44%

-0.18%

-0.56%

-6.06%

2/21/1997

801.77

0.85%
2.42%
1.28%
1.60%
1.37%

2.08%

2/24/1995

0.47%
0.48%
0.24%
0.55%

0.40%

-2.20%

-5.16%

2/14/1997

808.48

0.22%

-1.63%

-1.58%

-6.65%

7/18/1997

915.3

2.57%

3.48%

-1.58%

0.49%

6/6/1997

858.01

4.74%

6.32%

10.76%

3/16/1998

1079.3

1.51%

1.32%

3.38%

3.38%

7/3/1997

916.92

4.11%
0.03%

-0.18%

3.30%

-1.90%

4/27/1998

1086.5

3.27%

1.85%

0.69%

3.03%

63

3/2/1998

1047.7

4/20/1998

1123.7

1/25/1999

1234

5/10/1999

1340.3

7/19/1999

0.44%
3.30%

3.01%

4.38%

3.57%

2/1/1999

1273

-0.14%

-1.59%

-3.21%

5/17/1999

1339.5

2.30%
2.45%
2.28%
2.41%

-2.45%

-3.73%

2.18%

-3.38%

-2.86%

4.38%

0.20%

-0.30%

-3.45%

-5.06%

-4.82%

4.95%

3.39%

-3.79%

0.65%

-6.94%

-3.96%

-5.67%

1.98%
2.07%

-1.65%

2.36%

1.55%

-1.11%

0.52%

-0.11%

3.39%
0.38%
1.96%
5.65%
0.49%

4.24%

5.82%

-0.25%

-0.57%

-4.84%

-6.46%

-0.45%

2.15%

-1.82%

-6.25%

-3.89%

-6.90%

2.53%

-2.76%

-0.01%

4.17%
0.57%
3.86%

-0.17%

1.70%

3.66%

1.88%

1.91%

-5.44%

-9.97%

0.29%
12.17%

5.12%
0.84%

1.52%

0.55%

-5.50%

-1.84%

-4.64%

-8.31%

3.71%

6.60%

6.59%

-3.83%

-6.66%

-2.25%

-2.94%

-3.92%

3.79%

-3.17%

-6.19%

-5.54%

0.79%

3.01%

2.28%

8/2/1999

1328.1

-2.51%

-1.71%

4.15%

2/7/2000

1424.2

1407.7

3.16%
0.06%
4.25%

-5.65%

-5.46%

-5.07%

3/20/2000

1456.6

11/29/1999

1407.8

1.10%

0.52%

3.54%

-0.27%

4/3/2000

1506

3/13/2000

1383.6

10.14%

8.73%

2.06%

5/8/2000

1424.2

3/27/2000

1523.9

5.28%
1.17%

-1.27%

-3.05%

-9.84%

6/19/2000

1486

4/24/2000

1429.9

2.68%

-0.40%

-2.04%

3.22%

7/31/2000

1430.8

6/12/2000

1446

2.77%

0.64%

2.41%

2.54%

9/18/2000

1444.5

7/10/2000

1475.6

2.36%

-0.77%

0.25%

2.13%

10/9/2000

1402

8/14/2000

1491.6

1.51%

-0.64%

-7.01%

11/6/2000

1432.2

10/2/2000

1436.2

0.53%
2.38%

-4.29%

-2.62%

-6.97%

11/20/2000

1342.6

10/30/2000

1398.7

-3.39%

-4.47%

-4.99%

12/4/2000

1325

11/13/2000

1351.3

-1.12%

1.47%

-1.81%

12/18/2000

1322.7

11/27/2000

1349

2.32%

-2.50%

0.63%

2/12/2001

1330.3

12/11/2000

1380.2

2.40%
0.64%
1.78%
4.16%
1.77%

-4.71%

-4.85%

5/14/2001

1248.9

-5.57%

-7.42%

-3.45%
18.30%

5/25/2001

1277.9

1.12%
3.42%
0.01%
0.14%
1.57%

-0.04%

1.48%

-2.62%

10/29/2001

1078.3

-2.23%

-7.64%

-8.01%

1/4/2002

1172.5

-1.07%

2.60%

4.29%

1/18/2002

1127.6

-0.54%

-0.71%

-1.06%

3/28/2002

1147.4

0.49%
2.15%

-1.09%

-2.94%

1.74%

12/13/2002

889.48

0.70%

-1.13%

4.74%

-7.96%

-1.34%

-3.35%

-9.59%

5/2/2003

930.08

0.36%

1.53%

3.97%

4.78%

4.62%
0.10%

2/5/2001

1354.3

4/30/2001

1249.5

5/21/2001

1312.8

10/15/2001

1090

11/19/2001

1151.1

1/11/2002

1145.6

3/8/2002

1164.3

12/6/2002

912.23

0.16%
2.49%

-1.81%

1.17%

-7.52%

5/23/2003

933.22

3.62%

4.58%

5.19%

5.88%

4/25/2003

898.81

3.48%

3.85%

3.83%

9.22%

7/3/2003

985.7

1.26%

0.77%

-0.56%

2.26%

5/9/2003

933.41

1.17%

-0.02%

4.56%

7.97%

10/3/2003

1029.9

0.80%

0.92%

2.03%

3.91%

5/30/2003

963.59

2.60%

1.31%

3.42%

2/6/2004

1142.8

0.27%

-0.15%

0.39%

0.68%

6/20/2003

995.69

2.51%
1.96%

0.88%

-1.70%

0.41%

2/27/2004

1144.9

-2.13%

-3.22%

-0.82%

8/22/2003

993.06

1.51%

3.88%

3.00%

5.20%

4/23/2004

1140.6

-3.67%

-4.12%

-0.54%

10/10/2003

1038.1

-0.88%

1.46%

3.01%

6/18/2004

1135

-0.85%

-3.01%

-4.91%

2/13/2004

1145.8

0.12%
0.42%

0.89%

-3.61%

-1.43%

7/16/2004

1101.4

1.04%
2.92%
0.05%
1.38%

0.03%

-3.32%

2.22%

64

2.28%
2.30%

0.30%
0.13%
0.80%
1.24%

-0.52%

-2.13%

-1.81%

10/15/2004

1108.2

1.12%

1.99%

6.86%

8.16%

-0.18%

-1.95%

-6.27%

11/26/2004

1182.7

0.72%

0.45%

1.88%

-1.20%

-1.77%

-4.44%

-3.42%

12/10/2004

1188

1.42%

0.19%

1.20%

-2.35%

3.92%

6.07%

1/7/2005

1186.2

0.52%
0.14%

-1.89%

1.31%

2.80%

0.36%

2.06%

1.88%

7/22/2005

1233.7

0.04%

-0.59%

-1.13%

-0.22%

1191.2

1.54%
0.27%

0.26%

0.92%

-0.15%

8/5/2005

1226.4

0.32%

-0.55%

-0.68%

0.02%

12/17/2004

1194.2

0.90%

0.66%

0.15%

1.33%

12/9/2005

1259.4

0.63%

0.74%

2.41%

0.50%

7/15/2005

1227.9

0.51%

0.20%

1.03%

1/20/2006

1261.5

1.76%

0.20%

2.04%

3.46%

7/29/2005

1234.2

0.47%
0.63%

-0.31%

-2.36%

-1.50%

5/19/2006

1267

1.04%

-0.14%

-2.12%

-2.38%

11/11/2005

1234.7

1.84%

2.08%

4.82%

11/3/2006

1364.3

1.22%

2.70%

3.29%

3.96%

1/13/2006

1287.6

-0.19%

-1.92%

0.77%

1/26/2007

1422.2

1.84%

1.12%

1.91%

1.08%

5/12/2006

1291.2

-0.86%

-4.25%

-1.45%

3/2/2007

1387.2

1.13%

-0.02%

2.43%

6.86%

10/20/2006

1368.6

-0.31%

2.38%

3.94%

10/19/2007

1500.6

-2.79%

1430.5

1.25%

1.75%

-1.99%

5/16/2008

1425.4

-2.78%

-4.58%

-3.65%
14.76%

2/23/2007

1451.2

-3.33%

-1.04%

2.05%

11/19/2010

1199.7

2.31%
3.47%
1.00%

0.60%

1/19/2007

1.95%

3.95%

6.85%

10/12/2007

1561.8

-1.70%

-6.92%

-2.94%

12/17/2010

1243.9

1.10%

2.25%

4.11%

6.76%

5/2/2008

1413.9

0.64%
0.58%
4.41%
3.92%
1.81%

0.81%

-2.00%

-9.47%

12/31/2010

1257.6

2.83%

2.26%

3.87%

11/12/2010

1199.2

0.04%

-0.95%

3.44%

6.28%

1/21/2011

1283.4

1.10%
0.55%

2.14%

4.65%

1.17%

12/10/2010

1240.4

0.28%

1.38%

2.37%

6.79%

2/1/2013

1513.2

0.31%

0.44%

0.80%

3.77%

12/23/2010

1256.8

0.07%

1.17%

2.71%

4.67%

1/3/2014

1831.4

0.60%

0.40%

-4.89%

2.32%

1/7/2011

1271.5

1.71%

1.52%

3.74%

3.96%

1/25/2013

1503

1.00%

-1.01%

3.24%

12/27/2013

1841.4

0.68%
0.79%

-1.21%

-2.66%

0.20%

4/2/2004

1141.8

6/10/2004

1136.5

6/25/2004

1134.4

10/8/2004

1122.1

11/5/2004

1166.2

12/3/2004

1.10%
1.85%
1.87%

Bullish %: Bottom Decile Test (For Bullish Signals)

Bullish - Cross
% Neg

38%

35%

Bullish - Cross Back

32%

25%

% Pos

62%

65%

68%

75%

Net Ret

29%

44%

129%

182%

spx

1W

2W

4W

8W

12/4/1987

223.92

11.27%

15.50%

14.13%

12/31/1987

247.08

5.09%
1.49%

2.01%

4.04%

8.39%

1/22/1988

246.5

4.29%

1.81%

7.76%

4/15/1988

259.77

0.14%

0.60%

-1.15%

Date

% Neg

40%

28%

% Pos

60%

72%

Net Ret

23%

60%

spx

1W

2W

12/18/1987

249.16

-1.44%

3.80%

1/8/1988

243.4

3.55%

1.27%

9.02%

5/6/1988

257.48

-0.27%

-1.73%

4.49%

5/27/1988

253.42

5.38%

7.11%

65

Date

25%
75%
122%
4W
0.06%

25%
75%
170%
8W
4.03%

3.11%
3.72%
6.17%

9.85%
7.12%
4.64%

5/13/1988

256.78

9/2/1988

264.48

11/11/1988

267.92

11/25/1988

267.23

3/10/1989

1.46%

-1.31%

5.71%

4.31%

8/26/1988

259.68

1.85%

2.61%

0.75%
0.54%

1.64%

2.61%

5.48%

11/18/1988

266.47

0.81%

3.17%

0.27%

3.21%

5.26%

3/17/1989

292.69

-0.72%

1.26%

3.67%

3.98%

7.96%

4/28/1989

309.64

-0.66%

1.36%

292.88

1.71%
0.06%

-0.79%

1.44%

4.48%

7/21/1989

335.9

1.86%

2.39%

3/31/1989

294.87

0.78%

2.20%

5.01%

9.06%

9/29/1989

349.15

2.76%

-4.45%

7/14/1989

331.84

1.22%

3.11%

3.89%

4.77%

10/13/1989

333.62

4.06%

0.43%

9/22/1989

347.05

3.38%

0.03%

-1.57%

11/24/1989

343.97

1.94%

1.37%

10/6/1989

358.78

0.61%
7.01%

-3.24%

-5.90%

-2.05%

2/23/1990

324.15

3.51%

4.25%

11/17/1989

341.61

1.17%

2.87%

0.61%

-1.23%

3/23/1990

337.22

0.81%

0.85%

1/26/1990

325.8

2.40%

0.88%

3.63%

5/11/1990

352

0.75%

0.73%

3/16/1990

341.91

1.57%
1.37%

-0.58%

0.83%

3.76%

9/7/1990

323.4

-2.03%

-3.74%

5/4/1990

338.39

9/14/1990

316.83

12/21/1990

3.54%
4.67%

8.70%
7.53%

3.09%
3.86%

8.02%
5.48%

3.02%
-4.04%
1.64%
1.00%

3.22%
-1.01%
4.48%
-3.59%

4.03%
-1.83%
2.74%

2.13%
6.16%
1.28%

4.80%

8.57%

6.25%

12/28/1990

328.72

-4.04%

-4.94%

-3.40%

-5.30%

-0.98%

2/1/1991

343.05

4.75%

7.58%

7.66% 10.63%

331.75

4.02%
1.74%
0.46%

-3.68% -3.57%
2.22% 10.37%

-5.08%

-1.04%

10.06%

12/13/1991

384.47

0.67%

7.98%

1/11/1991

315.23

5.39%

6.61%

14.00%

18.31%

7/17/1992

415.62

-0.97%

2.07%

12/6/1991

379.09

1.42%

2.10%

10.11%

9.17%

8/7/1992

418.88

0.25%

-0.96%

7/2/1992

411.77

0.93%

3.02%

0.75%

8/28/1992

414.84

0.54%

2.51%

7/31/1992

424.21

0.69%
1.26%

-1.01%

-2.21%

-1.79%

10/2/1992

410.47

-1.90%

0.31%

9/4/1992

417.08

1.96%

1.21%

-2.28%

1.36%

10/30/1992

418.68

-0.26%

0.90%

10/9/1992

402.66

2.25%

2.84%

3.71%

8.11%

1/15/1993

437.15

-0.24%

0.37%

1/8/1993

429.05

1.89%

1.65%

4.63%

5.98%

6/4/1993

450.06

-0.62%

-1.42%

4/30/1993

440.19

-0.14%

2.27%

2.65%

6/25/1993

447.6

-0.39%

0.31%

6/11/1993

447.26

0.48%
0.80%

0.08%

0.38%

0.77%

8/27/1993

460.54

0.17%

0.33%

7/2/1993

445.84

0.70%

0.04%

0.97%

3.60%

3/18/1994

471.06

-2.22%

-6.82%

8/20/1993

456.16

0.96%

1.14%

-0.24%

2.69%

3/31/1994

445.77

0.30%

0.09%

3/11/1994

466.44

-1.26%

-3.55%

-4.38%

5/20/1994

454.92

0.53%

0.87%

3/25/1994

460.58

0.99%
4.70%

-2.33%

-1.71%

-1.25%

7/15/1994

454.16

-0.23%

0.90%

5/13/1994

444.14

2.43%

2.97%

3.37%

0.86%

8/5/1994

457.09

1.06%

1.44%

7/8/1994

449.55

0.79%

1.68%

4.77%

9/4/1998

973.89

5.73%

5.13%

7/29/1994

458.26

1.03%
0.26%

1.71% 2.65%
3.04% 1.02%
1.51% 14.14%

3.39%

0.56%

10/26/1998

1072.32

3.66%

5.40%

10.81% 12.24%

8/17/1998

1083.67

0.41%

0.80%
11.66%

-4.24%

-8.20%

7/16/2001

1202.45

-0.95%

0.17%

10/12/1998

997.71

6.48%

7.48%

13.28%

18.41%

10/4/2002

800.58

7/9/2001

1198.78

-0.65%

0.14%

-5.49%

4/1/2005

1172.92

9/27/2002

827.37

0.96%

8.49%

12.47%

4/22/2005

1152.12

0.41%

1.67%

3.23%

5.55%

1/24/2003

861.4

0.31%
3.24%
0.66%

-0.93% ######
12.54%
16.73%
4.34% 10.47%
2.20%
0.71% -2.58% -1.37%

-3.68%

-3.35%

0.33%

7/28/2006

1278.55

0.06%

-0.92%

1.29%

3.74%

66

9.36%
1.03%
-0.43%
0.43%

7.62%
2.32%
-2.70%
0.80%

2.00%
3.03%
1.70%
-0.94%
0.33%

5.09%
4.61%
3.27%
0.02%
1.70%

0.27%
-6.07%
1.63%
0.12%

0.79%
-4.60%
2.41%
-0.23%

3.07%

-0.86%
-4.59%
6.28%
2.74%

-8.46%
-1.81%
4.62%
-4.03%

-2.25%

-1.84%

-2.99%

-2.47%

-6.04%
15.24%

-2.30% ######

826.84

-4.52%
10.10%

3/13/2009

756.55

1.58%

11/13/2009

1093.48

-0.19%

0.20%

1.89%

4.77%

-4.64%

7/16/2010

1064.88

3.55%

3.45%

1.35%

5.35%

-0.20%

8.10%

9/3/2010

1104.51

1.57%

2.06%

3.17%

6.34%

6/3/2011

1300.16

-2.24%

5.49%

4.76%

10.10%

2.58%

6/17/2011

1271.5

-0.24%

3.75%
3.38%

5.38%

7.29%

11.37%

9/30/2011

1131.42

2.12%

2.94%
3.04%
-2.20%
2.67%
5.36%
8.23% 13.58%

7.23%
-1.02%
-5.27%
5.40%

-4.45%

-3.83%

5/25/2012

1317.82

-3.01%

-0.67%

-0.31%

1.55%

0.04%
0.44%
4.30%

-0.20%

3.82%

0.06%
11.92%

7/27/2012

1385.97

0.36%

1.44%

1.82%

5.12%

1.67%

8.96%

6.97%

4/19/2013

1555.25

1.74%

3.80%

7.22%

5.39%

-2.63%

-3.29%

-0.88%

1.71%
2.11%

2.08%

4.07%

7.23%

-0.42%

2.82%

3.40%

3/24/2005

1171.42

0.13%

0.83%

-1.65%

1.52%

11/30/2007

1481.14

1.59%

-0.89%

4/15/2005

1142.62

0.83%

1.25%

1.00%

5.09%

2/1/2008

1395.42

-4.60%

-3.26%

7/21/2006

1240.29

3.08%

3.15%

5.00%

6.52%

3/28/2008

1315.22

4.20%

1.34%

11/23/2007

1440.7

4.44%

3.04%

-7.09%

7/25/2008

1257.76

0.20%

1/4/2008

1411.63

-6.12%

-2.18%

-6.01%

11/28/2008

896.24

3/7/2008

1293.37

4.37%

6.12%

8.82%

1/9/2009

890.35

7/3/2008

1262.9

2.81%
0.75%
0.40%
1.85%

-0.18%

-0.21%

1.58%

2/13/2009

11/21/2008

800.03

13.71%

1/2/2009

931.8

2.02%
4.45%

3.43%
25.27%

1/30/2009

825.88

2/20/2009

770.05

5.17%
4.54%

0.12%
11.26%

8.95%
11.41%
15.14%

11/6/2009

1069.3

2.26%

7/2/2010

1022.58

8/27/2010

1064.59

5/27/2011

1331.1

6/10/2011

1270.98

9/23/2011

1136.43

5/11/2012

1353.39

7/20/2012

1362.66

4/12/2013

1588.85

-8.77%

-8.82% 1.77%
7.85% 13.51% 20.18%

3.46%

Intermediate Term Results

Bearish %: Top Decile Test (For Bullish Signals)
Bullish - Cross

Date

% Neg

37%

29%

Bullish - Cross Back

37%

31%

% Pos

63%

71%

63%

69%

Net Ret

33%

95%

108%

266%

spx

1M

2M

3M

6M

Date

% Neg

30%

26%

29%

30%

% Pos

70%

74%

71%

70%

Net Ret

99%

173%

200%

319%

spx

1M

2M

3M

6M

6/23/1989

328

1.73%

3.86%

5.66%

7.55%

6/30/1989

317.98

8.84%

10.73%

8.29%

7.82%

7/14/1989

331.84

3.89%

4.77%

8.43%

7.18%

7/21/1989

335.9

3.02%

3.22%

2.07%

3.76%

11/17/1989

341.61

0.61%

-1.23%

-2.81%

3.04%

11/24/1989

343.97

1.00%

-3.59%

-4.74%

3.05%

1/19/1990

339.15

-1.90%

1.29%

1.63%

7.75%

1/26/1990

325.8

0.88%

3.63%

1.40%

12.13%

2/9/1990

333.62

1.51%

2.32%

2.51%

5.35%

2/16/1990

332.72

3.25%

3.59%

6.48%

5/4/1990

338.39

8.57%

6.25%

5.25%

-7.62%

5/11/1990

352

2.74%

1.28%

-4.88%

2.17%
13.64%

67

8/3/1990

344.86

-6.47%

-8.68%

12.46%

-2.92%

12/7/1990

327.75

-3.92%

7.17%

14.77%

18.94%

12/21/1990

331.75

-1.04%

10.06%

10.90%

15.23%

12/28/1990

328.72

2.22%

10.37%

14.47%

14.21%

1/11/1991

315.23

14.00%

18.31%

18.50%

18.43%

2/1/1991

343.05

7.66%

10.63%

9.42%

11.05%

12/6/1991

379.09

10.11%

9.17%

7.98%

9.57%

12/13/1991

384.47

9.36%

7.62%

5.09%

7.55%

5/21/1993

445.84

0.09%

0.33%

1.64%

4.01%

10/30/1992

418.68

3.03%

4.61%

4.64%

4.38%

12/3/1993

464.89

0.12%

3.60%

-0.10%

-1.68%

5/28/1993

450.19

0.37%

-0.43%

2.13%

3.25%

4/8/1994

447.1

-1.07%

2.48%

-0.22%

3.39%

12/10/1993

463.93

2.44%

1.69%

0.42%

-0.82%

8/24/1998

1088.14

-5.38%

-2.22%

4.70%

14.13%

4/15/1994

446.18

-0.38%

3.63%

0.57%

1.39%

4/3/2000

1505.97

-3.96%

-5.67%

-3.80%

8/31/1998

957.28

9.58%

11.29%

23.58%

32.79%

12/18/2000

1322.74

1.91%

0.29%

-3.40%
13.02%

-5.06%

4/10/2000

1504.46

-6.14%

-2.20%

-4.52%

3/19/2001

1170.81

1.79%

6.72%

6.05%

-7.26%

12/22/2000

1305.97

4.47%

-4.07%

-3.18%
14.43%

9/17/2001

1038.77

4.93%

7.66%

12.47%

3/26/2001

1152.69

4.93%

13.59%

6.11%

6/28/2002

989.81

-9.18%

-4.23%

9.43%
17.23%

-9.97%

9/24/2001

1003.45

8.62%

14.71%

16.15%

7/19/2002

847.76

5.11%

-0.75%

9.42%

7/12/2002

921.39

-1.38%

-2.00%

9/6/2002

893.92

9.56%
10.44%

13.90%
14.77%

0.79%

4.54%

-5.90%

8/2/2002

864.24

6.00%

-5.67%

3.01%

-0.33%

9/27/2002

827.37

8.49%

12.47%

8.46%

8.27%

9/13/2002

889.81

-6.12%

0.55%

0.25%

-6.85%

1/31/2003

855.7

-2.44%

-0.88%

7.26%

14.71%

10/18/2002

884.39

2.88%

2.94%

3.83%

-1.82%

2/21/2003

848.17

5.61%

5.17%

8.56%

16.02%

2/14/2003

834.89

3.34%

6.03%

12.87%

16.68%

3/14/2003

833.27

4.20%

13.42%

18.19%

23.37%

3/7/2003

828.89

6.03%

11.78%

17.21%

20.26%

4/1/2005

1172.92

-1.37%

2.20%

1.52%

3.18%

3/21/2003

895.79

-0.42%

2.79%

12.93%

13.47%

5/6/2005

1171.35

2.23%

2.87%

6.21%

1.71%

4/8/2005

1181.2

-0.83%

1.38%

2.01%

3.02%

10/14/2005

1186.57

4.06%

6.22%

9.07%

9.18%

5/13/2005

1154.05

4.05%

5.91%

6.70%

5.26%

10/28/2005

1198.41

4.93%

4.85%

6.29%

9.15%

10/21/2005

1179.59

5.82%

6.81%

8.94%

8.96%

5/26/2006

1280.16

-2.31%

-0.88%

1.46%

9.09%

11/4/2005

1220.14

3.44%

4.37%

4.16%

6.97%

7/21/2006

1240.29

5.00%

6.52%

10.38%

15.35%

6/23/2006

1244.5

1.32%

4.26%

5.93%

13.56%

-7.47%
14.59%

-1.34%

8/4/2006

1279.36

2.47%

4.06%

7.70%

11.04%

7/28/2006

1278.55

1.29%

3.74%

7.70%

11.29%

12/1/2006

1396.71

1.42%

3.52%

-0.68%

8.69%

8/11/2006

1266.74

2.59%

6.62%

8.92%

14.23%

3/9/2007

1402.85

2.98%

7.60%

9.13%

4.34%

12/8/2006

1409.84

0.36%

2.73%

-0.50%

8.59%

5/4/2007

1505.62

2.23%

0.92%

-3.34%

0.68%

3/23/2007

1436.11

3.12%

6.20%

6.80%

3.33%

6/1/2007

1536.34

-2.15%

-4.06%

-4.53%

-6.29%

5/11/2007

1505.85

0.22%

0.28%

-1.94%

2.89%

8/17/2007

1445.94

2.12%

7.11%

-0.47%

-7.93%

6/8/2007

1507.67

1.60%

-2.65%

-1.21%

-2.56%

8/31/2007

1473.99

4.96%

4.54%

-6.93%

8/24/2007

1479.37

2.59%

1.82%

-3.12%

-8.75%

10/26/2007

1535.28

-8.34%

-2.53%

-3.10%
11.93%

-9.58%

9/7/2007

1453.55

7.16%

3.86%

1.30%

-8.46%

11/9/2007

1453.7

4.28%

-3.07%

-3.18%

11/2/2007

1509.65

-2.47%

1484.46

-9.83%

-9.56%

-8.37%

12/7/2007

1504.66

-7.61%

-4.14%
11.84%

-8.68%
13.31%

-7.50%

12/21/2007

-8.03%
10.44%

2/22/2008

1353.11

-0.24%

2.59%

1349.99

-5.44%

-1.15%

3.93%

1390.33

2.52%

3/28/2008

1315.22

6.28%

4.62%

0.21%

5/30/2008

1400.38

-8.71%

-2.17%
11.85%

-4.97%
29.16%
38.65%

2/15/2008

4/18/2008

5.43%
12.62%

4/25/2008

1397.84

-1.57%

-5.71%

-8.64%

-6.22%
12.08%
35.05%

-9.54%

68

-7.91%

6/13/2008

1360.03

8/29/2008

1282.83

9/12/2008

1251.7

11/21/2008

800.03

1/2/2009

35.98%
42.06%
45.40%

-3.75%
14.37%

-8.47%

-0.20%

8.10%

-9.69%
13.75%
28.16%

-4.02%
33.82%
25.62%

-9.96%
33.60%
27.32%

931.8

8.95%
11.41%

3.43%
25.27%

1/16/2009

850.12

-7.17%

2/20/2009

770.05

6/6/2008

1360.68
1242.31

-7.96%
11.52%

-8.21%
22.02%

9/5/2008
10/17/2008

940.55

-7.15%

-7.65%

13.71%

12/12/2008

879.73

-0.90%

-1.24%

1/9/2009

890.35

-2.30%

-5.23%
19.18%

0.23%

3.41%

3/20/2009

768.54

8.31%

18.14%

30.62%

4/17/2009

869.6

-6.11%
34.30%
10.41%
14.66%

36.99%
40.83%
-8.70%
6.75%

-8.40%

0.68%

18.37%

18.66%

35.86%

1.53%

6.22%

4.17%

21.62%

4/9/2009

856.56

8.48%

9.64%

2.86%

23.41%

6/5/2009

940.09

-4.40%

6.65%

8.57%

17.61%

5/22/2009

887

0.68%

7.62%

11.57%

23.85%

7/24/2009

979.26

4.79%

8.72%

12.12%

17.28%

6/19/2009

921.23

3.25%

6.35%

14.26%

18.97%

9/4/2009

1016.4

2.37%

2.61%

9.10%

9.77%

8/28/2009

1028.93

3.31%

3.70%

7.51%

7.69%

10/2/2009

1025.21

1.07%

6.87%

9.85%

13.79%

9/25/2009

1044.38

3.37%

4.50%

6.67%

11.06%

11/13/2009

1093.48

1.89%

4.77%

-1.37%

6.06%

11/6/2009

1069.3

3.17%

6.34%

-0.58%

12.43%

6/4/2010

1064.88

-3.97%

5.73%

-1.50%

10.88%

5/28/2010

1089.41

-1.36%

2.24%

-3.45%

8.19%

7/16/2010

1064.88

1.35%

5.35%

9.43%

19.88%

7/9/2010

1077.96

4.05%

2.46%

5.48%

16.87%

7/30/2010

1101.6

-3.36%

3.68%

7.63%

16.22%

7/23/2010

1102.66

-2.81%

3.63%

7.44%

16.62%

9/3/2010

1104.51

2.94%

7.23%

6.88%

19.50%

8/27/2010

1064.59

7.29%

12.52%

25.91%

6/17/2011

1271.5

2.67%

-5.27%

-7.76%

-0.83%

6/10/2011

1270.98

3.82%

11.37%
11.92%

-8.32%

-1.89%

8/19/2011

1123.53

7.17%

6.88%

11.41%

19.50%

8/5/2011

1199.38

-2.12%

-8.35%

4.50%

9.75%

10/14/2011

1224.58

3.21%

0.97%

5.54%

12.87%

9/23/2011

1136.43

8.96%

6.97%

6.06%

23.56%

5/25/2012

1317.82

-0.31%

1.55%

7.24%

3.19%

5/18/2012

1295.22

3.83%

5.28%

8.39%

6.53%

6/15/2012

1342.84

0.80%

4.56%

6.76%

5.60%

6/8/2012

1325.66

2.02%

5.17%

5.98%

6.83%

7/6/2012

1354.68

2.68%

3.83%

6.63%

3.52%

6/29/2012

1362.16

1.70%

3.54%

5.83%

4.99%

10/26/2012

1411.94

-0.14%

0.44%

6.25%

11.82%

10/19/2012

1433.19

-3.17%

0.18%

3.68%

8.29%

11/23/2012

1409.15

1.49%

6.08%

6.62%

18.33%

11/16/2012

1359.88

5.18%

8.29%

11.88%

20.14%

4/26/2013

1582.24

4.26%

-0.58%

6.96%

8.80%

4/12/2013

1588.85

2.82%

3.40%

3.99%

6.61%

Bearish %: Bottom Decile Test (For Bearish Signals)

Bearish - Cross

Date

Bearish - Cross Back

% Neg

41%

45%

43%

34%

% Neg

40%

40%

35%

31%

% Pos

59%

55%

57%

66%

% Pos

60%

60%

65%

69%

Net Ret

41%

3%

63%

225%

Net Ret

27%

38%

37%

229%

spx

1M

2M

3M

6M

spx

1M

2M

3M

6M

21.98%
26.46%

23.08%
16.77%

16.20%
16.77%

1.42%
24.41%
14.33%

28.55%
22.06%
13.05%

23.70%
15.29%

5.33%

0.00%

5.55%

9/18/1987

314.86

10/9/1987

311.07

10.21%
19.50%

2/5/1988

250.96

6.54%

3.01%

4.80%

3/11/1988

264.94

1.97%

-3.17%

0.09%

Date

7/31/1987

318.66

9/25/1987

320.16

6.00%

10/16/1987

282.7

2.63%
22.47%
13.11%

-1.29%

2/12/1988

257.63

3.39%

69

-4.69%

6/3/1988

266.45

2.00%

2.16%

-1.55%

0.29%

3/31/1988

258.89

0.94%

-2.11%

3.93%

4.35%

7/1/1988

271.78

0.16%

-3.48%

-1.30%

2.06%

6/10/1988

271.26

-0.26%

-0.47%

-2.09%

0.90%

1/20/1989

286.63

3.53%

1.15%

6.76%

15.11%

7/15/1988

272.05

-3.49%

-2.05%

2.28%

2.93%

2/17/1989

296.76

-2.30%

3.12%

6.24%

17.35%

2/24/1989

287.13

1.20%

7.51%

12.14%

4/12/1990

344.34

2.22%

5.02%

3.53%

-9.57%

4/20/1990

335.12

5.82%

6.49%

9.67%

20.38%
10.36%

4/26/1991

379.02

-0.40%

-2.13%

0.11%

3.64%

5/3/1991

380.8

1.91%

-0.76%

1.55%

1.88%

6/7/1991

379.43

-0.39%

1.48%

3.35%

-0.39%

6/14/1991

382.29

0.03%

1.50%

0.59%

-0.58%

6/21/1991

377.75

1.36%

-0.34%

2.05%

-0.01%

6/28/1991

371.16

3.23%

6.11%

4.46%

3.32%

9/13/1991

383.59

-0.56%

2.42%

-1.39%

5.98%

9/20/1991

387.92

1.18%

-1.37%

-0.89%

4.12%

4/24/1992

409.02

1.22%

-1.37%

1.16%

0.09%

5/1/1992

412.53

1.16%

-0.87%

1.21%

0.76%

5/29/1992

415.35

-2.87%

-0.92%

-1.11%

0.94%

11/27/1992

430.16

2.09%

2.28%

2.49%

4.75%

12/11/1992

433.73

-0.62%

2.67%

5.21%

3.77%

1/8/1993

429.05

4.63%

5.98%

3.09%

5.01%

1/15/1993

437.15

1.70%

3.27%

2.76%

2.63%

1/22/1993

436.11

-0.20%

2.93%

2.06%

3.01%

12/23/1994

459.83

1.31%

5.49%

7.79%

17.40%

12/30/1994

459.27

2.01%

6.12%

9.72%

19.99%

2/17/1995

481.97

2.94%

4.86%

9.59%

15.66%

2/24/1995

488.11

2.63%

5.08%

7.28%

14.72%

3/3/1995

485.42

3.15%

5.94%

7.86%

14.77%

3/10/1995

489.57

3.44%

7.02%

9.39%

14.57%

1/19/1996

611.83

5.91%

6.67%

5.42%

5.53%

11/10/1995

592.72

4.52%

0.97%

9.65%

8.25%

3/22/1996

650.62

-0.42%

3.46%

1.76%

3.16%

2/2/1996

635.84

2.35%

2.81%

2.88%

-0.73%

4/26/1996

653.46

3.83%

2.36%

-4.07%

7.80%

4/4/1996

655.86

-2.17%

1.80%

3.05%

4.57%

5/10/1996

652.09

3.08%

0.41%

1.58%

7.49%

5/3/1996

641.63

4.06%

5.34%

-0.99%

10.23%

6/14/1996

665.85

-5.41%

-0.01%

-0.31%

11.90%

5/24/1996

678.51

-1.42%

-7.61%

-1.89%

7.76%

1/17/1997

776.17

4.16%

2.52%

-2.76%

17.73%

6/21/1996

666.84

-4.96%

-0.04%

2.41%

11.08%

1/31/1997

786.16

1.16%

-3.37%

1.00%

19.61%

1/24/1997

770.52

5.16%

2.64%

0.53%

20.91%

2/14/1997

808.48

-1.58%

-6.65%

3.05%

17.65%

2/7/1997

789.56

3.05%

-2.97%

4.84%

20.86%

5/23/1997

847.03

3.73%

10.27%

9.32%

6.96%

2/21/1997

801.77

-2.20%

-5.16%

3.93%

15.00%

6/27/1997

887.3

5.54%

3.70%

7.28%

8.82%

6/13/1997

893.27

2.81%

4.90%

4.52%

9.35%

9/5/1997

929.05

3.87%

-1.55%

4.92%

12.95%

7/11/1997

916.68

1.84%

1.58%

6.11%

5.86%

10/10/1997

966.98

-4.08%

1.59%

-0.31%

16.10%

10/3/1997

965.03

-5.22%

1.01%

0.60%

13.51%

10/31/1997

914.62

6.58%

6.15%

7.75%

18.80%

10/17/1997

944.16

-1.67%

2.04%

1.46%

17.53%

12/8/1997

982.37

-1.87%

2.15%

7.46%

11.28%

11/24/1997

946.67

-0.80%

1.15%

9.66%

17.20%

1/12/1998

939.21

8.50%

13.76%

17.30%

1/26/1998

956.95

7.69%

15.54%

18.14%

24.01%

4/20/1998

1123.65

-1.59%

-3.21%

4.55%

23.22%
14.61%

3/16/1998

1079.27

3.38%

3.38%

3.06%

-8.99%

6/1/1998

1090.98

4.35%

3.60%

0.17%

4.90%

4/27/1998

1086.54

0.69%

3.03%

-3.59%

11/23/1998

1188.21

1.29%

3.11%

7.06%

12.21%

6/8/1998

1115.72

3.49%

-3.91%

7.14%
10.89%

2/12/1999

1230.13

6.27%

9.73%

10.20%

6.79%

11/30/1998

1163.63

6.72%

8.74%

6.42%

12.29%

5/10/1999

1340.3

-1.71%

4.15%

-2.61%

0.16%

2/22/1999

1272.14

1.95%

2.68%

4.81%

2.05%

7/19/1999

1407.65

-5.46%

-5.07%

-6.72%

-0.30%

5/17/1999

1339.49

-2.86%

4.38%

-2.80%

1.73%

6.38%

9/3/1999

1357.24

-3.88%

-0.23%

2.33%

-1.76%

7/26/1999

1347.76

0.92%

-2.98%

-6.41%

7.56%

11/29/1999

1407.83

3.54%

-0.27%

-3.86%

-0.06%

9/13/1999

1344.13

-0.66%

2.45%

4.84%

4.84%

12/27/1999

1457.1

-3.23%

-6.62%

2.99%

0.51%

12/20/1999

1418.09

2.67%

-2.15%

2.85%

1.97%

3/13/2000

1383.62

8.73%

2.06%

6.34%

9.69%

3/20/2000

1456.63

-3.79%

0.65%

0.95%

2.60%

70

6/19/2000

1486

0.52%

-0.11%

-0.07%

6/26/2000

1455.31

1.32%

2.94%

-0.27%

-2.92%

-9.59%
13.25%

7/3/2000

1469.54

-2.14%

2.74%

7/24/2000

1464.29

2.40%

8/14/2000

1491.56

9/11/2000

7/17/2000

1510.49

-1.25%

-1.89%

-8.17%

-0.30%

-7.81%

-9.97%

7/31/2000

1430.83

5.82%

-0.25%

-0.64%

-7.01%

-4.00%

8/28/2000

1514.09

-5.74%

-7.66%

1489.26

-5.86%

-3.83%

-7.57%

9/25/2000

1439.03

-3.01%

-6.70%

-9.27%

10/2/2000

1436.23

-2.62%

-6.97%

-7.47%

10/16/2000

1374.62

-1.70%

-0.25%

-3.48%

10/30/2000

1398.66

-4.47%

-4.99%

-3.13%

-9.20%
16.64%
19.74%
13.53%

-2.29%
11.01%

11/6/2000

1432.19

-3.89%

-6.90%

-5.78%

-6.15%
15.53%
18.64%
17.24%
11.57%

11/13/2000

1351.26

1.47%

-2.70%

1324.97

1.70%

-4.31%

-7.42%

2/12/2001

1330.31

-9.97%

3.66%
12.17%

-6.85%

1354.31

-5.62%

5/7/2001

1263.51

1.59%

-2.30%

5/25/2001

1277.89

-4.64%

-8.31%

1236.71

-2.06%

-6.43%

7/9/2001

1198.78

0.14%

-9.44%
11.37%

-8.72%
10.58%

7/2/2001

-3.77%
14.99%

-6.66%
10.96%
16.12%

12/4/2000

2/5/2001

-1.81%
18.30%

-6.42%

-7.86%
14.04%

7/16/2001

1202.45

-0.93%

-6.08%
13.61%

-9.35%

-3.82%

7/23/2001

1191.03

-1.65%

-5.49%
15.75%

-2.80%

-8.49%

-4.38%

10/22/2001

1089.9

5.61%

4.86%

4.49%

3.53%

10/29/2001

1078.3

6.60%

6.59%

5.10%

2.10%

12/14/2001

1123.09

2.06%

-0.41%

2.67%

12/21/2001

1144.89

-1.46%

-5.58%

0.76%

2/15/2002

1104.18

5.56%

2.19%

2/22/2002

1089.84

5.40%

1.65%

3/8/2002

1164.31

-3.35%

-9.59%

3/28/2002

1147.39

-6.19%

-5.54%

0.19%
13.48%

-9.50%
15.62%
24.22%

11/29/2002

936.31

-6.08%

-7.68%

-0.62%
10.62%
10.58%

-8.22%
18.00%
21.17%
-0.33%

12/13/2002

889.48

4.74%

-7.96%

-6.47%

9.72%

5/9/2003

933.41

4.56%

7.97%

3.43%

12.29%

5/23/2003

933.22

5.19%

5.88%

7.41%

13.43%

5/30/2003

963.59

1.31%

3.42%

3.13%

7.32%

6/6/2003

987.76

1.69%

-0.50%

3.47%

6.69%

6/27/2003

976.22

2.08%

1.79%

5.41%

10.27%

7/3/2003

985.7

-0.56%

2.26%

2.12%

11.19%

8/1/2003

980.15

2.84%

2.70%

5.20%

16.47%

8/8/2003

977.59

5.53%

5.81%

8.33%

15.71%

8/22/2003

993.06

3.00%

5.20%

5.09%

15.38%

8/29/2003

1008.01

-0.14%

2.29%

4.37%

13.19%

9/5/2003

1021.39

0.83%

2.87%

4.77%

12.10%

9/19/2003

1036.3

0.29%

1.36%

3.06%

8.13%

10/17/2003

1039.32

1.06%

2.76%

8.77%

10.19%

11/21/2003

1035.28

5.57%

10.50%

10.51%

4.72%

12/5/2003

1061.5

5.86%

6.13%

8.80%

5.58%

12/12/2003

1074.14

4.38%

7.78%

3.04%

6.17%

12/26/2003

1095.89

4.39%

4.36%

-0.42%

3.14%

2/6/2004

1142.76

0.39%

0.68%

-2.03%

-3.59%

2/13/2004

1145.81

-3.61%

-1.43%

-4.40%

-7.14%

2/20/2004

1144.11

-3.00%

-0.72%

-5.25%

-7.07%

3/5/2004

1156.87

-1.30%

-3.40%

-3.08%

-4.48%

3/12/2004

1120.57

2.20%

-2.99%

1.93%

-0.20%

4/16/2004

1134.61

-3.43%

-0.23%

-2.04%

-0.35%

4/23/2004

1140.6

-4.12%

-0.54%

-4.10%

-3.27%

6/10/2004

1136.47

-1.95%

-6.27%

-1.33%

4.83%

6/18/2004

1135.02

-3.01%

-4.91%

-0.59%

4.21%

7/2/2004

1125.38

-1.67%

-2.33%

-1.36%

7.48%

7/9/2004

1112.81

-4.39%

0.07%

2.01%

9.04%

12/23/2004

1210.13

-3.83%

-2.15%

-3.17%

0.07%

1/7/2005

1186.19

1.31%

2.80%

-0.18%

0.43%

6/17/2005

1216.96

0.34%

1.39%

1.17%

3.32%

7/1/2005

1194.44

3.43%

1.49%

1.78%

5.72%

7/15/2005

1227.92

0.20%

1.03%

-3.31%

3.71%

7/22/2005

1233.68

-1.13%

-0.22%

-3.53%

4.25%

7/29/2005

1234.18

-2.36%

-1.50%

-2.82%

2.21%

8/5/2005

1226.42

-0.68%

0.02%

-1.58%

4.67%

11/25/2005

1268.25

0.03%

-0.28%

1.54%

-0.10%

12/2/2005

1265.08

0.29%

1.37%

1.90%

1.19%

12/23/2010

1256.77

2.71%

4.67%

2.94%

0.86%

12/31/2010

1257.64

2.26%

3.87%

4.91%

2.06%

71

1/7/2011

1271.5

3.74%

3.96%

4.81%

3.86%

1/14/2011

1293.24

3.02%

-0.88%

1.62%

3.91%

1/6/2012

1277.81

5.21%

5.13%

10.61%

4.01%

1/20/2012

1315.38

3.49%

7.17%

5.73%

1.47%

1/27/2012

1316.33

3.89%

7.61%

4.23%

4.57%

2/3/2012

1344.9

1.44%

5.51%

4.53%

1.12%

7/12/2013

1680.19

0.67%

-0.50%

-0.24%

9.03%

7/19/2013

1692.09

-2.14%

0.33%

1.07%

8.63%

10/25/2013

1759.77

2.56%

3.88%

3.90%

6.37%

11/1/2013

1761.64

2.23%

4.92%

1.85%

6.12%

12/27/2013

1841.4

-2.66%

0.20%

0.61%

6.60%

1/3/2014

1831.37

-4.89%

2.32%

2.96%

6.87%

72