Market Indicators - Investor Sentiment

Top 321. May 15, 2000 7:13 AM

» Rande - Thanks for the good commentary Gene.

Thanks for the good commentary Gene. "Adrift" is an excellent description for those who find themselves without a solid long-term game plan which is founded in reasonable expectations.

-- posted by Rande

Top 322. May 15, 2000 9:49 AM

» JenL_2 - Put your chips down

Great Job Gene!

You said....

So put your chips down and let the Big Wheel spin.

Good Luck to everyone in America's favorite casino.....Jen

-- posted by JenL_2

Top 323. May 16, 2000 1:46 PM

» Demogremlin - Sentiment Chart

I charted the sentiment indicator for the data for the last 20 months or so. If someone wants to provide me with earlier data, I'd be happy to post that too.

here's the url: http://students.washington.edu/lrgib/sen...

I hope it works for everyone. There are two charts and a table of data.

Panspar

-- posted by Demogremlin

Glen's Suite101.com Market Timing Model Components

Top 324. May 16, 2000 10:49 PM

» JenL_2 - Investor Sentiment Charts

Demo - Welcome to Suite 101!

Nice charts! Haven't tried posting a link to several charts in an excel workbook. Good job!

Kirk just posted a chart on the “BB” thread where he has weekly sentiment, and 4 wk MA plotted against the S&P 500, but only back to June ’98.:

Editorial Comment

Also Glen has a 1 yr investor sentiment an 4 wk MA vs S&P 500 graph on his webpage:

which is posted on our "Links" Page:

http://www.suite101.com/links.cfm/invest...

Also Rande - thanks for the link to the investor sentiment article at TSC. Hadn't seen that one. Good explanation of the sentiment indicator, with pros and cons.

.....Jen

-- posted by JenL_2

Top 325. May 17, 2000 1:14 PM

» Demogremlin - Data digging anyone?

Thank you for pointing me to those excellent sites. I wish I'd known y'all were here a year ago. It seems many of you are Bob Brinker fans. I am too -- actually, he's a little conservative for me, but that's no crime. I've taken his 4% rule and made it a 10% rule for myself. And I make more timing decisions than he does (based largely on HIS analysis!) But conservative isn't all bad. And he is speaking to a general audience.

I can tell already that this will prove to be a valuable place to throw ideas around.

HERE'S MY QUESTION:

Would others be interested in collecting the data necessary to extend the %bull/(%bull+%bear) charts back as far as we can? Or do you know where the data is already in table form? What would this require? Would it be as tedious as looking through back issues of Barrons week by week?

Maybe if several of us pitched in, we could get the data fairly quickly. It seems that an hour in the library might be enough to get a year's worth of data...another hour of data entry... and viola!

I'd gladly pitch in and compile a year's worth of data in a group project. Is anyone else game?

Or is there an easier way?

-Panspar

-- posted by Demogremlin

Chartcraft Investors Intelligence

Top 326. May 17, 2000 2:03 PM

» Rande - Here’s the website for the II guys (they charge after free trial

Here’s the website for the II guys (they charge after free trial):

And you might want to try Meir Statman at Santa Clara University ((408) 554-4147 Fax: (408) 554-4029 E-Mail: mstatman@mailer.scu.edu)
Statman bio and links

Speaking of Statman, the following article is a GREAT resource on investor sentiment and its usefulness. It’s LONG and it’s ACADEMIC, so be warned. But, if you take this stuff seriously at all, it’s must reading.

Financial Analysts Journal
May/June 1998
Vol. 54, No. 3 Pg. 63-72; ISSN: 0015-198X; CODEN: FIAJA4
Bullish or bearish?
By: Clarke, Roger G; Statman, Meir

The sentiment of newsletter writers, whether bullish or bearish, does not forecast future returns, but past returns and the volatility of those returns do affect sentiment. High returns over four-week periods are associated with a migration of newsletter writers from the bearish camp into the bullish camp. High returns over periods of 26 and 52 weeks are associated with "nervous bullishness"-a migration of newsletter writers from the bearish camp into both the bullish and the correction camps. High volatility, instead of scaring newsletter writers into bearishness, reduces the effects of both positive and negative returns on sentiment. Also, contrary to a popular hypothesis, the crash of 1987 had no significant effect on the pattern of forecasts.

Investors are often divided into two groupssmart "information" traders and not-so-smart "noise" traders. Noise traders commit cognitive errors as they make their financial decisions; information traders do not. Information traders make the right decisions, and in the long run, they win at the expense of noise traders. But who are the information traders and who are the noise traders? And how do they forecast stock returns? This article attempts to answer these questions.

The Issues

Noise traders lose because they use trading rules that degrade investment performance rather than improve it. Few investors pin the noise trader label on themselves; others call them that. Investors Intelligence, for example, compiles the forecasts of writers of investment newsletters about future stock returns into a Bullish Sentiment Index, which is promoted as a contrary indicator; that is, users of the index turn bullish as the Bullish Sentiment Index turns bearish. So, users of the Bullish Sentiment Index pin the label "noise trader" on writers of investment newsletters.

Solt and Statman (1988) found that writers of investment newsletters deserve their noise trader label; they found no statistically significant relationship between the sentiment of investment newsletter writers, bullish or bearish, and subsequent stock returns. Followers of newsletter writers degrade their investment performance in two ways. First, they pay money for newsletters whose advice is no better than a free toss of a coin. Second, they move away from the optimal trade-off between risk and return in their strategic asset allocation in favor of faulty tactical asset allocation.

The forecasts of newsletter writers are faulty. But how do they form their faulty forecasts? The nature of forecasts is important because forecast patterns affect returns, volatility, and trading volume. In the framework suggested by De Long, Shleifer, Summers, and Waldmann (1990), noise traders are positive-feedback traders, traders who forecast continuations of past returns. Shefrin and Statman (1994) showed that the results of De Long et al. depended crucially on whether noise traders are indeed positive-feedback traders. The effect is quite different if noise traders expect reversals rather than continuations of past returns. Moreover, the pattern of forecasts may change from continuations to reversals, perhaps as a result of dramatic, memorable events in the stock market.

The market crash of 1987 was dramatic and memorable, and it seems to have exerted a profound effect on the prices of stock index options. Option prices in the post-1987 period exhibit a "skew" consistent with a high likelihood of another crash. Rubinstein (1994) dubbed this phenomenon "crash-o-phobia." Did the crash of 1987 affect forecasts in other ways? One argument is that the quick recovery from the crash of October 1987 and the even quicker recovery from the mini-crash of October 1989 taught investors that declines in stock prices should be viewed as buying opportunities, not as beginnings of long bear markets.

We investigated the patterns of forecasts that investment newsletter writers make, the effect of historical returns and volatility on the patterns, and the effects of the crash of 1987.

The Sentiment Data

Our data for newsletter writer sentiment are from the Bullish Sentiment Index, which is included in Investors Intelligence, a publication of the
investment service company Chartcraft. A June 5, 1995, report in Investors Intelligence states that the Bullish Sentiment Index is based on the analysis of more than 140 independent (non-brokerage-house) stock market newsletters. Examples of newsletters analyzed are the Zweig Forecast, the Cabot Market Letter, and the Todd Market Timer. For more than 30 years, claims the report, Investors Intelligence has found the Bullish Sentiment Index to be a contrary opinion indicator; that is, high bullish sentiment, as measured by the index, indicates that the market is making a top.

Newsletter writers, like all investors, are influenced by many factors as they make their forecasts of future returns. For example, the Zweig Forecast newsletter, quoted by Investors Intelligence, considers the advance/decline (A/D) figures as one factor. The May 19,1995, newsletter noted that the A/ D figures were positive in 20 of the previous 22 weeks and related that factor to expectations about stock returns:

We went back to 1928 and checked out all cases when the weekly A/D was positive in even 17 out of 20 weeks, a mark we hit this time on April 14. There were only 11 such prior instances. After the 17-of-20 figure was struck, the S&P 500 Index gained 4.7 percent in the next 3 months, 7.6 percent in 6 months, and 13.6 percent in 12 months vs. the normal 12-month gain of just 5.2 percent. Indeed, since April 14, the S&P has already tacked on some 3.5 percent.

For The Chartist newsletter, bond yields and price-dividend ratios of the DJIA are factors that influence forecasts. In the June 1,1995, newsletter, as quoted by Investors Intelligence, they noted that, in retrospect, their recent forecasts had turned out to be too bearish:

We feel much like a weatherman telling everybody that it is about to rain any minute; but when we open the door and look outside, it's 85 degrees and not a cloud in the sky! Obviously, this market has gone against us. So much so, that in a recent Hotline we indicated our intention of going back to the drawing board in an attempt to find out what went wrong.... In the meantime, we will admit that the pressure to recommend stocks has been intense. However, the pressure we have felt has not been from within because, first and foremost, we never second-guess our indicators. We have been publishing The Chartist since 1969 and, quite frankly, this is not the first time that we have been out of sync with the market.

Investors Intelligence classifies newsletter writers as bullish, bearish, or expecting a correction. The last category consists of newsletter writers who expect a decline in stock prices soon but an increase in stock prices in the long term. Investors Intelligence measures bullish versus bearish sentiment in its Bullish Sentiment Index, which is the percentage of bullish newsletter writers relative to the sum of bullish and bearish newsletter writers. For example, on May 5, 1995, the S&P 500 was at 520.12 and the DJIA was at 4343.40. On that date, 40.3 percent of newsletter writers were bullish and 34.5 percent were bearish. The Bullish Sentiment Index thus stood at 53.9 percent. The Bullish Sentiment Index does not take forecasts of corrections into account, but Investors Intelligence reported that 25.2 percent of newsletter writers on that date expected a correction.

Investors Intelligence publishes the forecast data approximately one week after newsletter writers express their forecasts in their newsletters. Because our interest was in the forecasts of the newsletter writers, we used the newsletter issue dates in our study. The sentiment figures were compiled by Investors Intelligence every month in 1963, every two weeks from 1964 to 1969, and every week since then. Our analysis began with the forecasts of newsletter writers on February 7,1964, and ended with the forecast of newsletter writers on May 19, 1995.

Value of the Forecasts

Solt and Statman found, using the data through 1985, that the Bullish Sentiment Index is useless as a forecast, straight or contrary, of future stock returns. They found no statistically significant relationship between the index and subsequent stock returns. We extended the time period through 1995 and found that the passage of time since 1985 did nothing to improve the forecasting ability of newsletter writers. As Figure 1 and Table 1 show, we found no statistically significant relationship between the level of the Bullish Sentiment Index (BSI) and subsequent S&P 500 returns for nonoverlapping periods of 4 weeks, 26 weeks, or 52 weeks.

Investors Intelligence has argued that the BSI is useful as a contrary indicator when its level is very high or very low, but that argument is not
consistent with the evidence. When we compared total S&P 500 returns for the four-week periods following the top 30 percent BSI readings with the S&P 500 returns for the four-week periods following the bottom 30 percent BSI readings, we found no evidence of the BSI's usefulness as a contrary indicator. As Table 2 shows, the mean four-week return following the bottom BSI reading is 0.58 percent, which is indeed higher than the 0.46 percent mean return following the top BSI readings, but the standard deviations of returns within both the top and the bottom groups are so high that the difference between the two means is far from statistically significant. Results for the 26-week and 52-week periods were similar.

Forecast Patterns

Newsletter writers have pronounced opinions about the future direction of the stock market, even though no statistically significant relationship exists between their forecasts and future returns. Why don't they recognize that the patterns they see in stock prices are, in fact, illusory?

Investors and newsletter writers who perceive patterns in stock returns that are, in fact, random, are not alone. The tendency to identify patterns in random data has been observed in many settings. As Gilovich, Vallone, and Tversky (1985) noted, intuitive perceptions of randomness depart systematically from the laws of statistics that underlie randomness. People apparently expect that if a series is random, the essential characteristics of randomness will manifest themselves not only in large samples but also in small ones. So, when they find patterns in a small sample of the series, they reject the possibility that the series is random. For example, knowing that coin tosses, given a large series of tosses, generate roughly half heads and half tails, people expect short sequences of coin tosses to contain roughly half heads and half tails. Large deviations from the half-and-half proportions are common in short sequences of coin tosses, however; much more common in fact than in people's perceptions of fact. Tversky and Kahneman (1971) described these common perceptions as a belief in the "law of small numbers," an erroneous belief that the law of large numbers also applies to small numbers.

People see patterns in random series, but which patterns do they see? Do they predict continuations of past numbers, or do they predict reversals? Researchers have found that predictions are highly sensitive to context. For example, when told that a coin has come up heads in three tosses in a row, most people predict a reversal for the next toss (this prediction tendency is known as "gambler's fallacy"). When told that a basketball player has hit three baskets in a row, however, most people predict continuation, in the belief that the player has a "hot hand." Gilovich, Vallone, and Tversky found that basketball players, coaches, and fans believe in the "hot hand" in basketball even though series of basketball shots follow a random walk. Why do most people predict reversals in the case of coin tosses and predict continuations in the case of basketball shots? Gilovich, Vallone, and Tversky attributed differences between predictions of continuation and predictions of reversal to differences in perceptions about the underlying process in the series. For example, it is hard to imagine a credible factor that could create a link between successive coin tosses, but many factors-confidence, fatigue, and so one could create a link between successive shots by a basketball player.

(Graph Omitted)

Captioned as: Figure 1.

Do forecasters' perceptions about the underlying processes of stock markets lead them to predict reversals or continuations? Common investment proverbs provide no answers because they reflect diametrically opposite perceptions of the underlying process. For each proverb that implies one should expect reversals (e.g., "trees don't grow to the sky"), there is a proverb implying that continuations are the rule (e.g., "don't fight the tape").

There is no necessary link between being bullish or bearish and the tendency to forecast continuations or reversals. At equilibrium price, the quantity of stocks demanded is always equal to the quantity supplied and the number of bullish dollars that change hands is equal to the number of bearish dollars. The fact that some investors are bullish at each point and others are bearish does not imply that one group of investors is forecasting continuation of past stock returns while the other is forecasting reversal. It may be, for example, that all investors recognize that there is no link between past stock returns and future stock returns and they are bullish or bearish on the basis of other information, such as the state of the business cycle or measures of valuation. Still, past returns seem to affect forecasts of future returns. Experiments provide most of the evidence on perceptions of the link between past and future stock returns. The experiments demonstrate that perceptions of the process that underlies stock returns are highly sensitive to setting. In one experiment, Andreassen (1988) presented some subjects with a series of past levels of stock prices and presented other subjects with a series of past stock returns (i.e., changes in stock prices). Subjects provided with levels of stock prices traded as though they expected reversals in stock prices; subjects provided with stock returns traded as though they expected continuations.

In another study, Andreassen (1987) presented past returns to one group of subjects and presented the same returns to another group but added to the return data news stories-positive news accompanying positive returns and negative news accompanying negative returns. He reported that the tendency to forecast continuations was higher in the group that received news stories than in the group that received no news stories. Andreassen and Krause (1990) used another set of experiments to focus on the "salience" (i.e., visibility or conspicuousness) of stock returns and the effect of variance of stock returns on that salience. They found that high variance reduces the salience of returns because it obscures both high and low returns. The reduction in the salience of returns reduces the tendency to forecast continuations. Subjects who were presented with a period when the overall return was positive but this positive overall return was accompanied by high variance of returns tended to be less bullish than subjects presented with the same total returns over a period when the returns were accompanied by low variance within the period. Similarly, subjects presented with a series of negative returns accompanied by high variance were less bearish than subjects presented with the same overall negative returns but the returns were accompanied by low variance.

(Table Omitted)

Captioned as: Table 1.

(Table Omitted)

Captioned as: Table 2.

Investors in real markets have access to information that is not controlled as it is controlled in experiments. Investors in real markets have information on prices, returns, variances, and news stories. What affects the forecasts of investors in real markets?

Forecasts: Continuations or Reversals?

Research indicates that some investors expect continuation of past stock returns and some don't. De Bondt (1993) found that individual investors surveyed by the American Association of Individual Investors (AAII) forecast future stock returns as though they expect continuations of past stock returns. In another study, De Bondt (1991) found that the economists in the Livingston survey, however, forecast future stock returns as though they expect reversals of the past stock returns. The reversal forecasts of the Livingston economists, by the way, proved to be no better than the continuation forecasts of the AAII investors; De Bondt (1991) found no statistically significant relationship between the forecasts of the Livingston economists and subsequent stock returns.

Imagine newsletter writers who have watched the stock market go up for some time. Will they turn bullish or bearish? The categories of bullish and bearish are not sufficiently fine to capture all forecasts. Some newsletter writers might be "nervous bulls," forecasting a "correction." Such writers are bearish in their outlook for the short term but bullish for the longer term. Similarly, some newsletter writers who have watched the stock market go down for some time might be "nervous bears" forecasting a "bounce," bullish in their outlook for the short term but bearish for the longer term. Experiments by De Bondt (1991) are consistent with nervous bullishness and nervous bearishness. Subjects who predicted increases in stock prices acknowledged a high likelihood of a correction, and subjects who predicted decreases in stock prices acknowledged a high likelihood of a bounce. We find that newsletter writers, as a group, display nervous bullishness, although we lack the data to examine whether they also display nervous bearishness.

High returns over a four-week period are associated with a migration of newsletter writers from the bearish camp into the bullish camp, with almost no effect on the correction camp. This finding is consistent with forecasts of future stock returns as continuations of past returns. As Table 3 and Figure 2 indicate, a 1 percent increase in S&P 500 returns over a four-week period was associated with a 1.23 percentage point increase in the number of bulls. The increase in the number of bulls came almost entirely from a 1.18 percentage point decrease in the number of bears. Only a tiny portion, 0.05 percentage point, came from a decrease in the number of newsletter writers expecting a correction. (The sum of the three must, of course, be zero.) When longer time periods are considered, however, the pattern of forecasts is different.

High returns over 26-week and 52-week periods are associated with "nervous bullishness," a migration of newsletter writers out of the bearish camp not only into the bullish camp but also into the correction camp.

An increase of 1 percent in S&P 500 returns in a 26-week period is associated with a 0.30 percentage point increase in the number of bulls. That 1 percent increase in the S&P 500 returns is also associated with a 0.70 percentage point decrease in the number of bears. These two pieces of data are consistent with a picture of newsletter writers forecasting future returns as continuations of past returns. When the newsletter writers in the correction camp are considered, the picture changes. That 1 percent increase in the S&P 500 returns for a previous 26-week period is associated with nervous bullishness in the form of a 0.40 percentage point increase in the number of newsletter writers expecting a correction.

The effect of returns for a previous 52 weeks on newsletter writers' forecasts of future returns is similar to the effect of returns of the
previous 26 weeks. High returns are associated with a migration of newsletter writers from the bearish camp into both the bullish camp and the correction camp. An increase of 1 percent in the S&P 500 returns is associated with a 0.60 percentage point decrease in the number of bears, a 0.24 percentage point increase in the number of bulls, and a 0.36 percentage point increase in the number of newsletter writers expecting a correction.

(Table Omitted)

Captioned as: ct Q

(Chart Omitted)

Captioned as: Figure 2.

(Table Omitted)

Captioned as: Table 4.

In sum, high returns for the short period are associated with increased bullishness, decreased bearishness, and almost no change in expectations for a correction. In contrast, high returns for the longer periods are associated with nervous bullishness: Bullishness increases and bearishness decreases, as in the four-week case, but a significant portion of newsletter writers migrate to the camp of those who, although bullish for the long run, expect a correction in the short run.

Effect of Volatility

Given that high returns over a four-week period drive newsletter writers out of the bearish camp into the bullish camp, is the drive also affected by the volatility of returns? One hypothesis is that, when returns are held constant, high volatility scares investors into bearishness. To test the hypothesis, we regressed the change in the BSI on returns and on the volatility of returns. The dependent variable was the change in the BSI over a four week period, and the independent variables were the total return on the S&P 500 over the four-week period and the standard deviation of daily returns within the four-week period.

Volatility does not scare newsletter writers into bearishness. If anything, as Table 4 reveals, high volatility appears to be associated with an increase in bullishness. The coefficient of the standard deviation is positive, although it is not statistically significant.

If volatility does not scare newsletter writers into bearishness, does it affect newsletter writers in other ways? Andreassen and Krause found in their experiments on salience of stock returns that volatility obscures the visibility of returns and, therefore, reduces the effect of past returns on forecasts of future returns. Positive total returns over a period together with high daily volatility of returns within that period made subjects less bullish than the same positive total returns with low volatility. Similarly, negative total returns accompanied by high volatility made subjects less bearish than the same negative returns accompanied by low volatility. In sum, high volatility obscured the absolute magnitude of returns, positive or negative.

We set out to examine whether the Andreassen and Krause results hold in real-life settings. A test of the salience hypothesis called for a separation of periods into two groups: the group of four-week periods with negative returns and the group of four-week periods with positive returns. Consider, as before, a regression of the change in the BSI over four-week periods on total returns over the periods and the volatility of daily returns within the periods. The salience hypothesis predicts that the coefficient of standard deviation will be negative for the group of periods with positive S&P 500 returns because, with returns held constant, high volatility should obscure the magnitude of positive returns and their otherwise bullish effect on the forecasts of future returns. We found (see the middle rows in Table 4) that the coefficient of the standard deviation for the group of positive S&P 500 observations is indeed negative and is statistically significant, a finding consistent with the salience hypothesis.

The salience hypothesis also predicts that the coefficient of standard deviation will be positive for the group of periods with negative S&P 500 returns because, with returns held constant, high volatility should obscure the magnitude of negative returns and their otherwise bearish effect on the forecasts of future returns. We found (see the last rows in Table 4) a positive and statistically significant coefficient of the standard deviation, a finding that is also consistent with the salience hypothesis.

In sum, volatility does not scare newsletter writers into bearishness. Instead, volatility obscures both positive and negative returns.

Effect of the Crash of 1987 on Forecasts

The argument that the quick recoveries from the crash of October 1987 and the mini-crash of October 1989 drastically changed the way investors forecast stock returns posits that before October 1987, bull markets were long and so were bear markets; therefore, forecasting continuations of past returns made sense.

However, the argument goes, the rapid recoveries after 1987 and 1989 taught investors that declines in stock prices should be regarded as buying opportunities, not as beginnings of protracted bear markets. For example, McGee (1997) wrote: Three weeks ago, when the Japanese Prime Minister Ryutaro Hashimoto rattled U.S. markets with comments that some analysts interpreted as a veiled threat to dump Japanese holdings of U.S. bonds and stocks, the industrial average plunged 192 points, its biggest point-size decline since "Black Monday" in the 1987 correction. But even as the correction was unfolding, money managers with cash to put to work were rubbing their hands with glee, adding to some of their favorite holdings with each drop. The following day, the average regained the vast majority of its loss, rising 153.80 points. (p. Cl) The question we examined is whether newsletter writers, as this argument would suggest, changed their ways of forecasting in the post-1989 period. We compared the relationship between changes in the BSI and S&P 500 returns for the 71 four-week observations in the post-1989 period (1990 through 1995) with the relationship for the 71 four-week observations in the pre-October 1987 period (1982 through 1987). As Table 5 shows, we found the relationship between the changes in the BSI and S&P 500 returns for both periods to be positive and statistically significant. The implication is that, on the whole, newsletter writers did not change their ways of forecasting; they did not see declines in stock prices as buying opportunities either before or after the crash of 1987. Moreover, a Chow test revealed the coefficient of the post-1989 period to be not statistically different from the coefficient of the pre-1987 period.

We found earlier that the pattern of forecasting varies with the period of past returns, so we also investigated whether the crash of 1987 affected the ways newsletter writers forecasted when post-crash periods were longer than four weeks. We did not have enough post-crash data for an examination of 26-week or 52-week periods, but an examination of 12-week periods, as shown in Table 5, indicated that the pattern of forecasts after the crash of 1987 was not different from the pattern before the crash.

In addition to studying changes in bullish/ bearish predictions between pre-and post-crash periods, which can be observed from the BSI data, we also wanted to study changes in writers forecasting a correction, which the BSI does not include but which Investors Intelligence does report. We found that the crash of 1987 did not affect the pattern of forecasts of corrections. A Chow test revealed no statistically significant difference between the patterns of correction forecasts before the crash and after it.

In sum, we found that the crash of 1987 exerted no statistically significant effect on the pattern of newsletter writers' forecasts.

Conclusion

From our study of the forecasts of investment newsletters, we conclude that the writers are unable to reliably forecast stock returns. We found no statistically significant relationship between the forecasts of newsletter writers and subsequent stock returns. We also found that newsletter writers cannot be classified merely as forecasting continuations of stock returns or as forecasting reversals.

High stock returns are associated with a move of some newsletter writers from the bearish camp into the bullish camp, which is consistent with the forecasting of stock returns as continuations of past returns. But high stock returns are also associated with a move of some newsletter writers into the correction camp, which is consistent with forecasts of stock returns as reversals of past returns. Time affects the pattern of movement. High returns over four-week periods are associated with a move from the bearish camp into the bullish camp; the effects on the correction camp are minor. High returns over 26week and 52-week periods are associated with major moves into the bullish and correction camps and away from the bearish camp.

Past stock returns are not the only factor that affects forecasts of stock returns. Volatility of stock returns also matters. We found that volatility does not scare newsletter writers into bearishness. Rather, volatility affects the sentiment of newsletter writers through its effect on the salience, or visibility, of returns: High volatility reduces the salience of both positive and negative returns. Thus, positive total returns over a period accompanied by high volatility within the period make newsletter writers less bullish than the same total positive returns accompanied by low volatility. Similarly, negative returns accompanied by high volatility make newsletter writers less bearish than the same negative returns accompanied by low volatility.

As for the argument that the rapid recoveries from the crash of 1987 and the mini-crash of 1989 taught investors that declines in stock prices should be regarded as buying opportunities, not as beginnings of bear markets, we found no effect of the 1987 crash on the pattern of newsletter writers' forecasts. No statistically significant difference exists between the pattern of forecasts before the 1987 crash and the pattern of forecasts after the 1987 crash.

Our study of the forecasts of writers of investment newsletters contributes to the body of work on the accuracy and influence of stock market forecasts. Earlier work by De Bondt (1991, 1993) examined the forecasts of economists in the Livingston survey and the forecasts of members of the AAII. Lee, Shleifer, and Thaler (1991) studied "small investors" who invest in closed-end funds, and Bernstein and Pradhuman (1994) studied brokerage firm professionals. Further studies of the forecasts of investors would be useful in contributing not only to our understanding of the behavior of investors in financial markets but also to our understanding of the process by which the interactions of investors in financial markets determine stock returns, volatility, and volume of trading.

We thank Jennifer Riehl and Roopa Trivedi for assistance and Peter Bernstein, Richard Bernstein, Michele LaPlante, Hersh Shefrin, and Steven Thorley for comments. Meir Statman acknowledges support from the Dean Witter Foundation.

(Table Omitted)

References omitted

Author Affiliation:

Roger G. Clarke is chair of Analytic TSA, and Meir Statman is a professor in the finance department at Santa Clara University.

-- posted by Rande

Suite 101 Market Timing Model

Top 327. May 17, 2000 2:23 PM

» JenL_2 - Great Demo!

You're just the person we've been looking for. A while back Glen started this thread...

....with the idea of collecting collecting, graphing, evaluating different market indicators, and to possibly come up with our own timing model. Well good idea - but Glen ended up doing all the work himself, so the project was tabled. So if you're interested in doing some "data digging" - Fantastic! Just read though the posts in the Suite 101 Market Timing Model thread and contact Glen by clicking on his email address in his profile. We can use your help Demo (Panspar). Welcome to the group!.....Jen

-- posted by JenL_2

Top 328. May 17, 2000 3:23 PM

» Rande - Here’s some excerpts for those who don’t want to wade through th

Here’s some excerpts for those who don’t want to wade through the whole thing (even though you should if you take investing seriously). If you've ever considered the "long-wave" stuff, the bit on randomness should be interesting to you:

Investors and newsletter writers who perceive patterns in stock returns that are, in fact, random, are not alone. The tendency to identify patterns in random data has been observed in many settings.

Conclusion

A study of the forecasts of investors is important for two reasons. First, the forecasts affect stock prices, volatility, and trading volume. Second, the trades that investors make based on their forecasts affect the welfare of the traders themselves. From our study of the forecasts of investment newsletters, we conclude that the writers are unable to reliably forecast stock returns. We found no statistically significant relationship between the forecasts of newsletter writers and subsequent stock returns.

-- posted by Rande

Top 329. May 17, 2000 7:03 PM

» Rande - Must add.

Must add....whether the sentiment numbers as meaningful indicator is just another in a long line of practical jokes perpetrated on unknowledgeable investors or not, one thing is sure to this investor -- Gene's commentary is something to look forward to. Now THAT'S "investor intelligence!"

-- posted by Rande