Forecasters of the stock market

Manage
1

From here [quote]
[size=3]Does Accurate Forecasting Get Attention?
[/size]
July 21, 2011 • Posted in Individual Gurus, Investing Expertise

Do individual experts whose U.S. stock market forecasting records are good (bad) gain (lose) attention? The “pro” argument is that investors (and online intermediaries) eventually flock to good forecasters and ignore bad ones in search of a market timing edge. The “con” arguments are that loud noise (for example, marketing-related or entertainment-driven) swamps information, and/or investors do not or cannot measure forecaster accuracy, and/or investors are more interested in ideas than forecasts. As a simple test these arguments, we compare two data series: (1) the stock market forecasting accuracies of gurus in the Guru Grades summary table; and, (2) the attention paid to these same individuals as measured by the number of search results found by a Google query on (“[guru name]” “stock market”), with the “stock market” qualifier intended to filter out potential namesakes and connect each name to the forecasted variable. Using results from searches for 60 individually graded gurus on 7/20/11, we find that:

The following scatter plot relates attention (number of Google search results on a logarithmic scale) to U.S. stock market forecasting accuracy for 60 individually graded gurus. The Pearson correlation for the two series is -0.05 and the R-squared statistic is 0.00, indicating no material relationship between attention and forecasting accuracy.

For another perspective, consider average guru forecasting accuracy by range of guru fame.

The next chart summarizes average guru forecasting accuracy for three ranges of guru fame as measured by number of Google search results. Average accuracy is consistent across ranges, again indicating that famous gurus are no more or less accurate than obscure ones.

For a third perspective, the fame-weighted average accuracy of all 60 gurus is 46.4%, compared to a raw average of 48.0%. Given the sample size, these averages are not materially different.

In summary, evidence from simple tests on a limited sample indicates no relationship between the forecasting accuracy of and the attention paid to U.S. stock market gurus. It seems that other factors drive investor attention, which tends to concentrate on a very few gurus.

Cautions regarding findings include:

As noted, sample and subsample sizes are small.
The significance of this analysis depends on the reliability of judgments in Guru Grades regarding guru forecasting accuracies and on the meaningfulness of a definition of attention based on Google search results.
Google search results for a given guru can vary considerably over time, with variation perhaps deriving from news spikes or changes in search technology.

[/quote]

2

Thanks for posting the article. When I read articles written by the Gurus it is for entertainment value only. Years ago I realized that their crystal ball works no better than mine, and often worse.

Scott

3

Interesting that people pay no more attention to the guru’s who have been right than to those that have not. I guess it’s all about marketing. What I find even more informative is that the average guru is right less than 50% of the time and the distribution of guru accuracy is quite like a bell curve: http://www.cxoadvisory.com/gurus/. There are ways to make money in the markets, but trying to predict the future is not one of them.

4

That’s right, it’s all about marketing. Forecasters (and the media/blogs in general) don’t sell by being right. They sell by attracting attention. See here. The worst investment year I ever had was when I tuned in to the news.

Some of Warren Buffet’s quotes comes to mind:
“We’ve long felt that the only value of stock forecasters is to make fortune tellers look good…”
“Forecasts may tell you a great deal about the forecaster; they tell you nothing about the future.”

While they may not be so famous, there are forecasters that are better than average. David Tepper has has been a step ahead of everyone else at least since the beginning of the financial crisis. His hedge fund made so much money year after year that it looks like typos.

One of the few forecasters tracked by cxoadvisory who has been accurate is Ken Fisher. He invented the PSR in the early 80’s. He correctly called and profited from the dot-com crash. While he completely missed the financial crisis misreading his own signals, I went to cash in 07 due to Ken Fisher’s market forecasting techniques. Based mostly on Ken Fisher’s book it became obvious to me in August of 07 that it was time to sell. Which I did. Building on the experience that he shared in the book and using his techniques I called the bottom in 09 on the day after it happened. I was 100% invested in 2009 for the only time since 2006.

The following were some signs of the 07 market top that were taken from Ken Fisher’s book:

  1. Financials were 40% of the S & P 500 by market cap. (Energy reached 40% in the early 80’s, technology did in the dot-com crash)
  2. A nine month range where the market made sort of a double top; ie the market hit a high and then couldn’t surpass it for nine months.

The following signs of the 2009 market bottom were in Ken Fishers book:

  1. A double bottom three months apart.
  2. A tapering off of volume indicating less forced liquidations
  3. A big daily upwards spike in daily volume and price
5

I think people forget that the market itself is an aggregate forecast of all market participants. This is a “Wisdom of the Crowd” effect.

In addition the time and resources required to accumulate enough information to make such a forecast probably means that it is not going to be broadcast for free.

6

Don, the fact that the average guru is right less than 50% of the time does not prove that there are no gurus who consistently time the market accurately. To say otherwise would mean that it is impossible to generate alpha because 50% beat the market and 50% didn’t beat the market.

7

There are gurus who consistently accurately time the market but it is very hard to gain access to them because they do not need your money. Renaissance Technology is an example. They rarely open funds and when they do they have very high minimums and close quickly. The free forecasting advice mostly is worth what you pay for it.

Chipper, I am glad Fisher’s book helped you to make money. There are many books that unfortunately help people to loose money.

Scott

8

One of my favorite quotes

Niels Bohr “Prediction is very difficult, especially if it’s about the future.”

Jim

9

All true. But I was lucky. Unlike many people, I knew before I started investing that most gurus are not reliable. I have a background in disproving theories that emanate from ivory towers and have learned to be quite skeptical of “experts”. Therefore I chose Fisher only after studying his track record and his experience and read the book with a critical eye. I didn’t follow Fisher blindly either. (Fisher himself was 100% long during the financial crisis and never claimed to be infallible). Instead I studied the historical stock market data that he presented in the book and the anecdotes that he wrote about from his many years of experience in the market (Warren Buffett studied his father’s techniques). I learned his techniques for sniffing out what works and what doesn’t.

10

Nobody knows anything - is what I have concluded.
Your own forecast is as good or better as those from the “experts”.

Check out Albert Edwards. His procrastination are dutifully reported by the media, but they are totally wrong. Skeptics and devotees of technical analysis took notice in July 2012 when Albert Edwards, the closely followed investment strategist at Societe Generale, warned the S&P 500 was “on the verge of an ultimate death cross,” foretelling imminent major losses for the stock market, forecasting the S&P would collapse to 666, and “all hope will be lost”. When I checked this I found that the ultimate death cross was actually a harbinger of good fortune, and I was proven to be correct by the subsequent performance of the market. http://advisorperspectives.com/newsletters12/The_Ultimate_Death_Cross.php

Or Hussman, who has been telling us for the last 5 years that stocks are overvalued, only to see his own fund implode. Also there is Lakahman Achuthan from the ECRI who has been telling us since September 2011 and in subsequent countless TV appearances that we are in a recession, and still holds to this believe. Another one is David Rosenfeld. One should not believe anything these guys come up with - it is all a lot of nonsense.
Georg

11

Hussman, Rosenberg, Lakahman, Pretcher, etc. are all brilliant guys but they probably lack the ability to recognize they did a mistake. They have something in mind and they will find all facts supporting their belief. The facts they find can be incredible and convincing but they fail to look at the big picture and adapt. I’m not saying they cannot be right, but in some cases they will lose all their business or be bankrupt before they get proven right. If they are ever right, the mass see them as the ones who cried for the wolf but nobody believes them anymore they lost all credibility. The only reason they are still in business is for the contrarian opinion, that even if wrong at the moment, at least they made their DD about what could go wrong more than anyone else. To take with a grain of salt.

12

The history of market forecasters is a long and disasterous route. I go back to Joe Granville and Elaine Garzarelli who had tremendous followings back in the day but eventually lost their following due to bad calls. In the eighties Robert Prechter and Garzarelli developed very big followings after they “called” the '87 market crash. But looking into their logic, you’d see huge flaws. Prechter actually associated the building of skyscrapers in large metro cities to signs of a stock market top. Whereas Garzarelli had developed a quant based approach which worked for a time before it didn’t.

The other issue that must be stood is that “forecasters” and strategists were big money makers for their respective firms, which was in my view their reason for existence. They were the stars, the important talking heads and they brought in clients and attention. But look no further than Abbey Joseph Cohen as the primo example of useless “strategic guidance”

The role of strategists appears to be evolving with the decline of the big brokerage firms. The good ones don’t try as much to call market tops or bottoms like Doug Kass (one good call in a decade) , but they offer views that are frequently quant based that offer some guidance as to which asset class or sector might be the next to top or bottom.

Charlie Bilello - he’s new and on Twitter and his views are technically based but he speaks mostly in terms of sector guidance. He spotted the bottom in Emerging markets a few weeks ago by taking a cue from EM credit spreads.

Richard Bernstein - formerly from MLynch and on his own wrote a book a decade ago “Navigate the Noise” which endorsed quant strategies as a method of success. He doesn’t pick tops or bottoms and has been endorsing the small cap run for many years.

13

I understand, but the average combined with a bell curve comparable to what would be expected from completely random predictions shows that most forecasts probably have little value. I don’t equate trading (reading current market signals, assessing probabilities, controlling risk on the trade, exiting when you are wrong/or taking profits when you are right) with forecasting.

14

Following these gurus and “forecasters” is a sure way to lose money. Yes, sometimes they are right and that creates so much makreting hype that people follow them blindly (“he called the crash of…”). Most of the time they are totally wrong and their “prediction” is as good as yours or mine.

The market is a chaotic beast and there are so many variables and factors influencing it that noone can predict anything with any consistent accuracy.
Yes, you can make money in the markets. But listening to and relying on “predictions” is definitely not one of them.

15

And the lesson for us all is, market timing does NOT work!!!

We can bet the market, but not by timing it.
Instead we must ride the right factors/ranking systems consitently through up and down markets. Forget timing, hedging, …

“Now, I’m certain of success. I go down with the water and come up with the water. I follow it and forget myself. The only reason I survive is because I don’t struggle against the water’s superior power.” James O’Shaughnessy quoting a Taoist story.
:slight_smile:

16

Tobiasberr,

I agree with what you say. I find it interesting that you like Mandelbrot’s book. As I recall, his main point was that volatility in the market is not random and can to some extent be predicted. Of course, that does not translate to someone like me predicting the direction of the market.

I wonder what points you took from his book.

Thanks.

17

I’m going to try a completely different angle on this.

I suggest there is absolutely zero randomness in the market and in theory, that each and every movement is perfectly, reliably and consistently predictable. The reason we don’t experience this in the real world is because the data-points we’d need and the relevant relationships are so numerous, we can’t come anywhere near capturing all of them. Under this view, randomness would be defined as the aggregate of far more deterministic events than we can feasibly identify or measure.

But suppose after a few more centuries of continuing acceleration in progress in terms of data capture, storage, processing and analysis – way beyond anything any of us can envision today – is it not possible that market prediction cam ultimately become a simple, reliable, routine matter? It’s not as if it would take a sudden and explosive development to get us there. Most likely, it would only require continuing progress along an already-existing and well-entrenched trend along these lines.

Consider, for example, meteorology. We like to make fun of weather predictions today, but seriously, don’t even mediocre forecasters today have a far better grasp of things than those to whom this was important in, say, the middle ages? How about medicine: As imperfect as we still are, haven’t we come a long way from the days when everyone assumed disease could be cured by bloodletting? And consider even the dismal science itself, economics. Yeah, yeah everybody likes to bash whoever it is who happens to be chairing the Fed or living at the White House etc., but seriously, don’t you think our ability to pull up from a crash so much faster in 1987, 2008, etc. than was the case in 1929 and earlier has at least something to do with an improved understanding of relevant data-points and the way they relate to one another and how all the if-then scenarios are likely to play out? Meanwhile, quite a few in the past generation achieved much fame explaining how stock prices were unpredictable and that nobody could beat the market, yet aren’t there quite a few P123 members who’ve been doing just that by working with data-points in ways not conceivable by the efficient markets gurus? And consider Warren Buffett, the enemy of market timing referred to in an earlier post: Couldn’t we say that he has, over the course of many years, predicted (and backed his forecasts with real money) that whatever market mishaps might occur would not be sufficient to offset the impact of judicious stock selection? (That’s as much a market prediction as any other.)

Prediction is a viable endeavor and one at which the world is continually getting better. Eventually, we’ll probably reach perfection, but that’s for another future point in history, not the one in which we’re living. But if we want to do the best we can with what we have today, it would seem that we’d be better off trying to understand what we’re good at and what we’re not, and focusing as best we can on the former.

18

Its already true that if I want to know what is going to happen next year, you can probably tell me in about 100 days :slight_smile:

19

No. Predicting markets is nothing like predicting the weather. My prediction of the weather does not influence the weather. If I say there will be a hurricane tomorrow it cannot create a hurricane, nor influence one if it exists. If I say the market will go up tomorrow, then I may buy the market along with others who predicted the same, and so now our prediction has influenced the market. One has to not only predict what the market will do but what people that are predicting the market will do and what influence that will have on the markets … and how people will react to that and what influence that will have on the markets, and so on.

Markets are based on people and people are not always rational. That’s not to say that prediction cannot improve.

Totally agree.

20

dwpeters,
I totally agree, the feedbackloops make the system even more complex.

mgerstein,
you can beat the market by relying on certain rules/factors. This will increase your Chances.

But I do not believe, that the market is a deterministic system.
Because as a result, the world and the human brain would be deterministic. (at least quantum mechanics is not deterministic, which is part of our world)
I hope and want to believe, that our mind is free.
We might be predictable to some extent, or with certain probability.

But not to the last degree.
So in some instances, I hope, we make completely free and random decisions.

One could argue, that in aggreggate these decisions would even out, but they do not.

And this is where Mandelbrot comes into the equation. (@Jrinne)
The market is a chaotic system, not smooth randomness. (butterfly effect…)
It is not rolling a dice a million times (no laplace, no Black Scholes, …), it is far more wild and chaotic fractal.

At the end of the day trillions of small circumstances, some deterministic some not, interact in a complex system each day and some lead to huge effects that move the markets.
So we will have to ride the bear and bulls instead of trying to time them.