12 Cognitive Biases That Endanger Investors

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
richard
Posts: 7961
Joined: Tue Feb 20, 2007 2:38 pm
Contact:

12 Cognitive Biases That Endanger Investors

Post by richard »

Nice survey of cognitive biases, such as Confirmation Bias, In-Group Bias, Gambler’s Fallacy, etc. Try to guard against these behaviors.

http://www.minyanville.com/special-feat ... &refresh=1
User avatar
baw703916
Posts: 6681
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: 12 Cognitive Biases That Endanger Investors

Post by baw703916 »

That was a good list, thanks for posting it, Richard.

My comments on a couple of the items:

#2 (in-group bias) applies just as much on this board as anywhere else.

#3 (gambler's fallacy) strictly speaking only applies to a system with no memory. No reasonable person would claim that market participants don't remember what happened yesterday. Also because price moves change valuations, a completely random series of returns a la the roulette table would tend to make valuation ratios eventually become arbitrarily large or small (i.e. BtM of 135 or 0.06). Such things do happen, but generally only to individual companies under severe stress to the point where their survival is in doubt.
Most of my posts assume no behavioral errors.
pkcrafter
Posts: 15461
Joined: Sun Mar 04, 2007 11:19 am
Location: CA
Contact:

Re: 12 Cognitive Biases That Endanger Investors

Post by pkcrafter »

Thanks Richard. No question that behavioral mistakes are the No.1 reason for poor investor returns.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.
Wagnerjb
Posts: 7213
Joined: Mon Feb 19, 2007 7:44 pm
Location: Houston, Texas

Re: 12 Cognitive Biases That Endanger Investors

Post by Wagnerjb »

baw703916 wrote:#3 (gambler's fallacy) strictly speaking only applies to a system with no memory. No reasonable person would claim that market participants don't remember what happened yesterday. Also because price moves change valuations, a completely random series of returns a la the roulette table would tend to make valuation ratios eventually become arbitrarily large or small (i.e. BtM of 135 or 0.06). Such things do happen, but generally only to individual companies under severe stress to the point where their survival is in doubt.
Brad: this fallacy applies to people who believe there are patterns in stock market returns. Just like there are no patterns in gambling results, there are no patterns in stock returns. Market efficiency is the reason stock returns are essentially random, like gambling results. "What happened yesterday" is what reset the daily expectation for today's returns at the normal return. The market resets the price every day so that the expectation for tomorrow is the normal return. In that context, what happened yesterday is irrelevant to investing decisions today.

Best wishes.
Andy
User avatar
baw703916
Posts: 6681
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: 12 Cognitive Biases That Endanger Investors

Post by baw703916 »

Andy,

Are you claiming that momentum doesn't exist? How can it exist if the market is a memory-less system?

The market does have a memory. One example that applies even to people on this board: say the market goes up (or down) a substantial amount in the next few months. Will Bogleheads just shrug and ignore it? Well, hopefully not; they should look at their portfolio, consult their IPS as appropriate, and see whether rebalancing is called for. So their actions are determined by what the market does. If enough people do this, it will impact the market's behavior.

Best wishes,
Brad
Most of my posts assume no behavioral errors.
User avatar
nisiprius
Advisory Board
Posts: 52211
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: 12 Cognitive Biases That Endanger Investors

Post by nisiprius »

Personally, I'm not so sure that momentum does exist. For a good century now, investors have believed in the existence of momentum by that name. If it really ever existed, it's pretty hard for me to believe that nobody has caught on to it yet and arbitraged it away.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
baw703916
Posts: 6681
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: 12 Cognitive Biases That Endanger Investors

Post by baw703916 »

nisiprius wrote:Personally, I'm not so sure that momentum does exist. For a good century now, investors have believed in the existence of momentum by that name. If it really ever existed, it's pretty hard for me to believe that nobody has caught on to it yet and arbitraged it away.
If you try to arbitrage it away, you run the risk that the market can stay irrational longer than you can stay solvent. ;)
Most of my posts assume no behavioral errors.
connya
Posts: 132
Joined: Tue Mar 05, 2013 3:01 pm

Re: 12 Cognitive Biases That Endanger Investors

Post by connya »

I always thought momentum was a generally accepted sort of thing.

Isn't the use of tolerance bands pretty common here? Without momentum there does not seem to be any reason for them besides possibly lowering transaction costs.

There is a lot of discussion about it in this thread: http://www.bogleheads.org/forum/viewtop ... 10&t=10670 but there is also a lot of confusion.

If even the existence of momentum is questionable then I should probably just drop the concept of tolerance bands completely from my personal investing rules.
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 8:40 am

Re: 12 Cognitive Biases That Endanger Investors

Post by staythecourse »

Speaking of momentum the data is the data.

If you look at serial correlations from one year to another for large cap, small cap, corporate bonds, and government bonds they are a big fat zero. I believe cash has a mild positive of 0.3 to 0.4. That is more likely due to government intereference of settting interest rates as monetary policy.

This is a classic example of the human mind trying to find patterns when there are none there.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
allocator
Posts: 94
Joined: Sun Dec 23, 2012 11:10 am

Re: 12 Cognitive Biases That Endanger Investors

Post by allocator »

Personally, I'm not so sure that momentum does exist. For a good century now, investors have believed in the existence of momentum by that name. If it really ever existed, it's pretty hard for me to believe that nobody has caught on to it yet and arbitraged it away.
Couldn't the same be said of the value and small cap premiums?
mindbogle
Posts: 142
Joined: Sun Feb 10, 2013 10:28 am

Re: 12 Cognitive Biases That Endanger Investors

Post by mindbogle »

One interesting thing that I remember from my university stats class about the Gambler's Fallacy (or Gambler's Ruin as I think it was called, but I think they are related) is that in the coin toss experiment, the ratio of heads to tails converges to one as the number of flips approaches infinity, but the cumulative net number of wins (+1) and losses (-1) for a player diverges to infinity as the number of flips approaches infinity. So the longer you play, the bigger net winner or loser you are likely to become, and the longer (more flips) it takes to cross over from a net winner to a net loser, or visa versa.
User avatar
baw703916
Posts: 6681
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: 12 Cognitive Biases That Endanger Investors

Post by baw703916 »

mindbogle wrote:One interesting thing that I remember from my university stats class about the Gambler's Fallacy (or Gambler's Ruin as I think it was called, but I think they are related) is that in the coin toss experiment, the ratio of heads to tails converges to one as the number of flips approaches infinity, but the cumulative net number of wins (+1) and losses (-1) for a player diverges to infinity as the number of flips approaches infinity. So the longer you play, the bigger net winner or loser you are likely to become, and the longer (more flips) it takes to cross over from a net winner to a net loser, or visa versa.
I think this has a lot to do with the confusion over whether stocks become less risky or more risky when you hold them for a longer time period. The rate of return more closely approaches its expectation value as the holding period increases, but the range of final portfolio values increases.
Most of my posts assume no behavioral errors.
mindbogle
Posts: 142
Joined: Sun Feb 10, 2013 10:28 am

Re: 12 Cognitive Biases That Endanger Investors

Post by mindbogle »

staythecourse wrote:
If you look at serial correlations from one year to another for large cap, small cap, corporate bonds, and government bonds they are a big fat zero. I believe cash has a mild positive of 0.3 to 0.4. That is more likely due to government intereference of settting interest rates as monetary policy.
staythecourse, that's interesting, because when I looked at serial correlations for large and small cap stock 1-year returns over long windows, I found small negative correlations (on the order of -10%, plus or minus). I experimented with Monte Carlo simulations that simulate both cross- and serial correlation of multi-asset class portfolio returns and convinced myself that it only takes a small negative serial correlation of stock returns to affect simulation outcomes in a very significant way (significantly lowering the simulated outcome wealth dispersion). I also concluded that the impact of that small negative serial correlation in returns was consistent with the evidence of mean-reversion price behavior reported by Campbell and Shiller, and others.

Granted, investors shouldn't conclude (even based on a small negative serial correlation) that because the stock market went up this year, that is going down next, or vice versa. But I do think there is compelling statistical evidence of weakly mean reverting prices, and these maybe should be considered in asset allocation (but don't want to digress - covered in other threads!).

It seems to me less common for investors to fall for Gambler's fallacy bias (the idea that memory-less games mean-revert), and more common to fall into "recency bias." Did the authors miss that one?
User avatar
jeffyscott
Posts: 13484
Joined: Tue Feb 27, 2007 8:12 am

Re: 12 Cognitive Biases That Endanger Investors

Post by jeffyscott »

staythecourse wrote:Speaking of momentum the data is the data.

If you look at serial correlations from one year to another for large cap, small cap, corporate bonds, and government bonds they are a big fat zero. I believe cash has a mild positive of 0.3 to 0.4. That is more likely due to government intereference of settting interest rates as monetary policy.

This is a classic example of the human mind trying to find patterns when there are none there.

Good luck.
I thought the data does show momentum, perhaps over shorter term periods such that it would not be found by looking only at annual returns data. Ken French even has a momentum factor page: http://mba.tuck.dartmouth.edu/pages/fac ... actor.html
mindbogle
Posts: 142
Joined: Sun Feb 10, 2013 10:28 am

Re: 12 Cognitive Biases That Endanger Investors

Post by mindbogle »

mindbogle wrote:One interesting thing that I remember from my university stats class about the Gambler's Fallacy (or Gambler's Ruin as I think it was called, but I think they are related) is that in the coin toss experiment, the ratio of heads to tails converges to one as the number of flips approaches infinity, but the cumulative net number of wins (+1) and losses (-1) for a player diverges to infinity as the number of flips approaches infinity. So the longer you play, the bigger net winner or loser you are likely to become, and the longer (more flips) it takes to cross over from a net winner to a net loser, or visa versa.
baw703916 wrote:
I think this has a lot to do with the confusion over whether stocks become less risky or more risky when you hold them for a longer time period. The rate of return more closely approaches its expectation value as the holding period increases, but the range of final portfolio values increases.
I think you are right. Successive compounding (multiplying as opposed to summing) magnifies the divergence effect.
User avatar
VictoriaF
Posts: 20122
Joined: Tue Feb 27, 2007 6:27 am
Location: Black Swan Lake

Re: 12 Cognitive Biases That Endanger Investors

Post by VictoriaF »

I think the article referenced in the OP is sloppily written, and several statements intermingle various issues in confusing or wrong ways. Specifically,

1. Confirmation bias
VF: The description starts by correctly describing the human propensity to look for the justification of one's opinions. But it ends with we always strive to see “both sides of every trade”. This is wrong. For example, if at a specific instance, one is subject to the confirmation bias that selling is the right thing to do, then one does not see the buyer's side of the trade. In other words, this bias is about seeing only your side, not both.

3. Gambler’s Fallacy
VF: I agree with Brad. Roulette outcomes are a truly random process along with the coin tosses and throws of dice. The markets are not random. Market participants have a memory of the market values the day before, in 1894 and every day in between.

4. Post-Purchase Rationalization
VF: The post-purchase part seems to be a version of the endowment effect. Whatever is ours (things, purchased securities) is treasured higher than its intrinsic value for those who are only contemplating a similar purchase. Rationalization is a bias of its own. It applies not only to purchases but also to not making purchases, and various other acts.

5. Neglecting Probability
VF: This bias was extensively studied by Kahneman and Tversky and it specifically refers to base probabilities. A representative hypothetical question is something like the following. There are two hospitals in a town, one small, one large. In January, in one of the hospitals 40% girls and 60% boys were born; and in the other, there were 55% girls and 45% boys born. Which hospital is which? People usually respond that 40/60 happened in the large hospital. In fact, the larger the hospital is, the closer the probabilities are to 50/50.

The interpretation in the article of this bias as not to confuse brains with a bull market is wrong. There is no discernible base probability associated with the bull market.

7. Status-Quo Bias
VF: Most of the description is fairly accurate, but the last sentence is awful, Change isn’t only positive, it’s inevitable. Some change is positive, some change is not. Even if change is inevitable, it may still be possible to maintain the status quo.

Consider the Bogleheads mentality. There could be changes bringing in new types of assets or hedge funds. The Bogleheads, however, would still continue investing based on their IPS.

8. Negativity Bias
VF: What does the biased attraction to negativity have to do with the last clause, and there’s the added twist that the stock market is widely considered to be a leading indicator?

9. Bandwagon Effect
VF: Even if people are generally inclined to follow what the others do, there are more significant factors that cause fund managers to follow the herd. If a fund invested differently from others and under-performed, the manager would be fired. If a fund invested similarly to other funds, then all funds would under-perform as a group and the manager would keep his jobs by blaming the under-performance on the market.

11. The Current Moment Bias
VF: As described it seems to be Kahneman's and Tversky's time discounting bias.

12. Anchoring Effect
VF: This bias was also demonstrated by Kahneman and Tversky. The most shocking finding was how different the anchor could be from the number in question. In a famous experiment, the participants looked at the number displayed on a wheel of fortune and that number influenced their estimate of the number of countries in Africa. The article describes it much more specifically, i.e., looking up stock-related data to estimate its value. The way the article describes it, it's not a case of the "classical" anchoring and, in fact, looking up relevant data may be useful for estimating a stock's value.

In summary, while the understanding of behavioral economics and finance is very useful, the article linked in the OP provides a poor review of the topic.

Victoria
Last edited by VictoriaF on Sun Mar 17, 2013 10:35 pm, edited 2 times in total.
Inventor of the Bogleheads Secret Handshake | Winner of the 2015 Boglehead Contest. | Every joke has a bit of a joke. ... The rest is the truth. (Marat F)
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 8:40 am

Re: 12 Cognitive Biases That Endanger Investors

Post by staythecourse »

mindbogle,

I never crunched my own numbers, but David Darst in his book "art of asset allocation" and roger gibson in "asset allocation" (think it was those two books, but could have been a Mr. Ferri book thrown in there as well ?AAAA) has the charts supporting what I wrote.

jeffyscott,

It is my fault of using the term momentum too loosely. I meant the adjective and not the financial term. The data definetly shows no correlation postive (the adjective momentum) or correlation negative (RTM) from one year to the next. Thus the description of random walk.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
Wagnerjb
Posts: 7213
Joined: Mon Feb 19, 2007 7:44 pm
Location: Houston, Texas

Re: 12 Cognitive Biases That Endanger Investors

Post by Wagnerjb »

baw703916 wrote:
nisiprius wrote:Personally, I'm not so sure that momentum does exist. For a good century now, investors have believed in the existence of momentum by that name. If it really ever existed, it's pretty hard for me to believe that nobody has caught on to it yet and arbitraged it away.
If you try to arbitrage it away, you run the risk that the market can stay irrational longer than you can stay solvent. ;)
Brad: I should have been more specific in my reply. When we speak of a market being efficient, it means that there are no inefficiencies that can be profitably exploited. There certainly may be smaller efficiencies and patterns, but none that can be exploited for consistently abnormal returns. I am not sure that momentum can be consistently and profitably exploited (as you seem to suggest as well), so I put that into a category of patterns that don't matter.

Best wishes.
Andy
User avatar
jeffyscott
Posts: 13484
Joined: Tue Feb 27, 2007 8:12 am

Re: 12 Cognitive Biases That Endanger Investors

Post by jeffyscott »

staythecourse wrote:I meant the adjective and not the financial term. The data definetly shows no correlation postive (the adjective momentum) or correlation negative (RTM) from one year to the next.
I'm not sure what your distinction is supposed to mean, but...
Since the 1980s academic studies have repeatedly shown that, on average, shares that have performed well in the recent past continue to do so for some time. Longer-term studies have confirmed that this “momentum” effect has been observable for much of the past century.
http://www.economist.com/node/17848665
Professor Emeritus
Posts: 2628
Joined: Mon Aug 13, 2012 6:43 am

Re: 12 Cognitive Biases That Endanger Investors

Post by Professor Emeritus »

baw703916 wrote:Andy,

Are you claiming that momentum doesn't exist? How can it exist if the market is a memory-less system?

The market does have a memory. One example that applies even to people on this board: say the market goes up (or down) a substantial amount in the next few months. Will Bogleheads just shrug and ignore it? Well, hopefully not; they should look at their portfolio, consult their IPS as appropriate, and see whether rebalancing is called for. So their actions are determined by what the market does. If enough people do this, it will impact the market's behavior.

Best wishes,
Brad
You are confusing cause and effect. even if individual traders have memories it does nto mean that the market as whole has a memory. Given your example it is equally true to say that the market moves because traders anticipate reactions by other traders. Classic game theory without any claim of "market memory".
Post Reply