Serving the Quantitative Finance Community

 
User avatar
yabbadabba
Topic Author
Posts: 261
Joined: July 2nd, 2006, 5:35 pm

Behaviourial Finance Models

January 11th, 2007, 7:07 pm

At the moment I'm studying Shleifer's behavioural finance (BF) book. It has some nice arguments and calculations but nothing much I can use in the markets (of course). I've heard there are BF orientated funds, but haven't come accross any noteworthy so far.One very interesting aspect is that agents in financial markets are conservative. This means they update their belief to slowly. First they underreact and then they overreact. I'm wondering how one could map news flow and prices to sentiment.It seems to me BF orientated models could be best used in a longer time frame in the equity markets. There is a steady relatively isolated news flow.
 
User avatar
Maelo
Posts: 1243
Joined: July 28th, 2002, 3:17 am

Behaviourial Finance Models

January 12th, 2007, 9:22 am

There are some papers where the "change in mindset" is slow and it is modelled by a Markov Chain. At this very moment, I can't recall the authors; however, it is old stuff.As soon as I get a chance I'll post the reference for you.
 
User avatar
flairplay
Posts: 130
Joined: September 26th, 2006, 1:34 pm

Behaviourial Finance Models

January 13th, 2007, 9:57 am

The best way to understand the market is to know that it is full of dopes. Behavioural Finance is a fancy term for this.There are dopes in the cash business, in the derivatives business, there are dopes with PhD's, perhaps even Nobel's and so on. And I am not kidding either. Dont mistake the concept of dopiness here with general stupidity. Some dopes are academically smart, some not. But generally they are not really interested in understanding what is going on. My defintion of a dope is someone who has a tremendous need to believe what received authority and the majority says. So it all becomes self fulfilling dopiness.Being a dope is no impediment to doing well for yourself, as the other dopes will tend to be duped.Then there are the wannabe dopes who go around and read pseudo philosophical stuff written by self appointed experts. It's good for the smaller proportion who know they know little but that the best way is to keep your mind open.All these dopes dont necessarily do badly for themselves. In fact some have made tremendous money running hedge funds etc, but in the end using the pure yardstick of whether they got their trading right, they generally fail over an extended period of time.Non dopes, not necessarily smart themselves, sit around and observe with a mixture of bemusement and sadness, while refusing to be fooled.That is a trader take on Behavioural Finance.
Last edited by flairplay on January 12th, 2007, 11:00 pm, edited 1 time in total.
 
User avatar
yabbadabba
Topic Author
Posts: 261
Joined: July 2nd, 2006, 5:35 pm

Behaviourial Finance Models

January 13th, 2007, 2:04 pm

Well, yes you're right. The question is: who is fooled by randomness?BF introduces some vague ideas, but I, like everybody else (except the scientists), want to make money out of the market. My question therefore is: how do we get from observed biased behaviour to models which make profit? How do we get more information about the behaviour without making vague guesses all the time. BF makes the typical gross assumptions that economists have to make.For me there is no conceptual framework within which I can work in. There should be a mathematical trader theory.
Last edited by yabbadabba on January 12th, 2007, 11:00 pm, edited 1 time in total.
 
User avatar
mensa0
Posts: 191
Joined: January 20th, 2004, 8:56 am

Behaviourial Finance Models

January 14th, 2007, 5:47 am

I think the first main-stream article on overreaction was LeBondt and Thaler (JF, 1987) which documented non-trivial momentum (perseverance) in price series. In my opinion though, the modeling of the behavior of market participants, while interesting, is not going to make you money.Here's how I like to make money: My daughter wanted an IPod for Christmas '04. I went to three Best Buy stores in the Milwaukee area to find one in pink (no other color would do!) Not only were the pink ones sold out, but every other color was in very short supply. I finally found one at the fourth store I visited, and this was a full three weeks before Christmas.After going home and checking some fundamentals and analysts' estimates, I bought AAPL. Their second quarter report (March '05) was good news indeed (and a "surprise" on the Street!)Christmas '05, my daughter's wish list was clothes - all from American Eagle Oufitters. I went to the closest AEOS store to get her a gift card, again about three weeks before Christmas. The place was packed!! Teenagers, tatoos, piercings, loud incomprehensible music playing, etc., and a lot of their parent's credit cards. I bought AEOS the same week.I make the most money in stocks that I personally see something that I don't think the Wall Street analysts do.Merton Miller was asked about behavioral finance just before he died in 2000. His response: "Temporary anomalies in search of a theory." I agree with him.Mike
 
User avatar
yabbadabba
Topic Author
Posts: 261
Joined: July 2nd, 2006, 5:35 pm

Behaviourial Finance Models

January 14th, 2007, 12:07 pm

Nice story indeed. Unfortunately I'm a quant-trader and I build models. I could send a robot the stores through One simple example: conservatism. People update their internal model to late. That means one can easily trade stocks after the fact, i.e. news. To come back to the story: why not just scan for sales increase? Very often one finds stocks after they made a good return and thinks, well that's no surprise. So I agree, very often the easy things work.
Last edited by yabbadabba on January 13th, 2007, 11:00 pm, edited 1 time in total.
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

Behaviourial Finance Models

January 14th, 2007, 2:43 pm

Mensa0, you need to publish a hot-stock tips newsletter based on your daughter's tastes. Subscriber revenues might just cover the costs of your daughter's shopping sprees. The question is, does Mensa0's behavior appear in the price data? Can we spot a trend in which savy consumers are buying stock in hot retail companies before institutional buyers get wind of official company sales figures?But the bigger picture issue is whether quants will kill BF patterns. BF suggests that fear, greed, and slow human nearly-monotonic logic could create a pattern in the prices. This pattern would appear to be efficient to other humans who have a similar pattern of fear, greed, and bounded rationality.But what happens when emotionless, fast-logic traders (= computers) start trading. Initially, some of the computers might do no better than human traders because the computer's software is based on the same flawed human logic or human traders (displaying BF) might override computer trade signals based on flawed gut models. But, eventually, dispassionate speed would seem to be superior to emotional sloth.This gets us back to consumer-driven patterns in the prices (versus trader-driven or institutional-driven price patterns). Perhaps BF's future will contract to a subsegment of the market in which BF is used to predict the impact of retail trade on stock trades.
 
User avatar
KackToodles
Posts: 4100
Joined: August 28th, 2005, 10:46 pm

Behaviourial Finance Models

January 14th, 2007, 3:04 pm

QuoteOriginally posted by: mensa0Christmas '05, my daughter's wish list was clothes - all from American Eagle Oufitters. I went to the closest AEOS store to get her a gift card, again about three weeks before Christmas. The place was packed!! Teenagers, tatoos, piercings, loud incomprehensible music playing, etc., and a lot of their parent's credit cards. I bought AEOS the same week. I make the most money in stocks that I personally see something that I don't think the Wall Street analysts do.Why can't wall street analysts see that demand for ipods or AEO products is growing? This is precisely what analysts do -- track demand and same store growth. What you describe id NOT behavior -- it 's the very rational rise in stock prices in response to improving fundamentals! Behavioral has to do with IRRATIONAL behavior of stock prices. Or irrational posts to wilmott, ha ha.
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

Behaviourial Finance Models

January 14th, 2007, 3:48 pm

QuoteOriginally posted by: KackToodlesQuoteOriginally posted by: mensa0Christmas '05, my daughter's wish list was clothes - all from American Eagle Oufitters. I went to the closest AEOS store to get her a gift card, again about three weeks before Christmas. The place was packed!! Teenagers, tatoos, piercings, loud incomprehensible music playing, etc., and a lot of their parent's credit cards. I bought AEOS the same week. I make the most money in stocks that I personally see something that I don't think the Wall Street analysts do.Why can't wall street analysts see that demand for ipods or AEO products is growing? This is precisely what analysts do -- track demand and same store growth. What you describe id NOT behavior -- it 's the very rational rise in stock prices in response to improving fundamentals! Behavioral has to do with IRRATIONAL behavior of stock prices. Or irrational posts to wilmott, ha ha. First, retail comps are reported with a lag. A savy consumer will know that a store is crowded a week or more before an analyst does (unless the analyst visits the stores themselves). The consumer can still be wrong (the store's margins might suck due to excessive discounting which creates both extremely high sales but also high losses).But this phenomenon isn't just about fundamentals. There are stocks that are persistently overvalued on the fundamentals due to the popularity of the company. Consumers that like the products and like the company tend to buy the stock with less regard for the numerical value of inscrutable metrics such as P/E ratio etc. In such a situation, the price dynamic of the stock switches from being set by expectations of future cashflows to a function of the time-rate-of-change in popularity. (The fundamentals can't be bad, but as long as the company is profitable, it can be popular out of proportion to fundamental financial projections.) On one level of analysis, price=popularity is irrational. But the more appropriate term is bounded-rational because these consumer-investors are using a subset of information available to them to make buy/sell decisions.
Last edited by Traden4Alpha on January 13th, 2007, 11:00 pm, edited 1 time in total.
 
User avatar
gardener3
Posts: 1496
Joined: April 5th, 2004, 3:25 pm

Behaviourial Finance Models

January 14th, 2007, 5:10 pm

QuoteOriginally posted by: KackToodlesQuoteOriginally posted by: mensa0Christmas '05, my daughter's wish list was clothes - all from American Eagle Oufitters. I went to the closest AEOS store to get her a gift card, again about three weeks before Christmas. The place was packed!! Teenagers, tatoos, piercings, loud incomprehensible music playing, etc., and a lot of their parent's credit cards. I bought AEOS the same week. I make the most money in stocks that I personally see something that I don't think the Wall Street analysts do.Why can't wall street analysts see that demand for ipods or AEO products is growing? This is precisely what analysts do -- track demand and same store growth. What you describe id NOT behavior -- it 's the very rational rise in stock prices in response to improving fundamentals! Behavioral has to do with IRRATIONAL behavior of stock prices. Or irrational posts to wilmott, ha ha. We know how relaible the anlayst forecasts are.... Also, not all behavioral finance has to do irratinality, some deal with preferences. Saying loss aversion is irrational is like saying risk aversion is irrational
 
User avatar
KackToodles
Posts: 4100
Joined: August 28th, 2005, 10:46 pm

Behaviourial Finance Models

January 14th, 2007, 8:47 pm

QuoteOriginally posted by: gardener3We know how relaible the anlayst forecasts are....Nobody's talking about analysts forecast. we're talking about basic market research to show that sales are growing. is such market research less reliable than his daughter-ial annecdotes?
 
User avatar
gardener3
Posts: 1496
Joined: April 5th, 2004, 3:25 pm

Behaviourial Finance Models

January 15th, 2007, 3:50 am

Because analysts are very good at telling you what has already happened. You need more to get an edge in the market. If going to a mall, as Peter Lynch advocates for instance, helps you get this kind of information earlier than the market, then why not? Of course, this does not mean that historical information or the research analysts produce is useless. I have no prior as to the usefulness of the mensa's daughter anecdotes, but there is research that shows that the market reacts favourably to the release positive results from customer satisfaction surveys, and that companies whose products are rated well by the consumers have higher returns in the market. Again if somehow you can get a sense of this information earlier then others you'll have an edge. Reading the report after the price has gone up won't be of much use.
 
User avatar
KackToodles
Posts: 4100
Joined: August 28th, 2005, 10:46 pm

Behaviourial Finance Models

January 15th, 2007, 5:48 am

QuoteOriginally posted by: gardener3Because analysts are very good at telling you what has already happened. You need more to get an edge in the market. If going to a mall, as Peter Lynch advocates for instance, helps you get this kind of information earlier than the market,First, what you can see at the mall has, by definition, already ocurred so it is historical information. Second, field researchers also go to the mall. Third, investors also go to the mall. What's your point?
Last edited by KackToodles on January 14th, 2007, 11:00 pm, edited 1 time in total.
 
User avatar
yabbadabba
Topic Author
Posts: 261
Joined: July 2nd, 2006, 5:35 pm

Behaviourial Finance Models

January 15th, 2007, 8:11 am

QuoteOriginally posted by: Traden4AlphaBut this phenomenon isn't just about fundamentals. There are stocks that are persistently overvalued on the fundamentals due to the popularity of the company. Consumers that like the products and like the company tend to buy the stock with less regard for the numerical value of inscrutable metrics such as P/E ratio etc. In such a situation, the price dynamic of the stock switches from being set by expectations of future cashflows to a function of the time-rate-of-change in popularity. (The fundamentals can't be bad, but as long as the company is profitable, it can be popular out of proportion to fundamental financial projections.) On one level of analysis, price=popularity is irrational. But the more appropriate term is bounded-rational because these consumer-investors are using a subset of information available to them to make buy/sell decisions.That is a very interesting remark. You suggest, that are two different regimes, one in which the fundamentals are dominant and one in which there is an obvious shift to popularity. This is what Soros described as reflexivity and is quite different to the models in Shleifer's book. These two regimes should have notably different features (volume and momentum). To trade this kind of approach one needs both fundamentals and price information itself. I think technical analysis which is obviously not a scientific theory, aswell as economical theory has surprising little to say about these phenomena. BF might be on the right track, but it has a long way to go. What I'm wondering is which kind of quantitative methods investment or hedge funds use for fundamental equity strategies and for global macro strategies.
 
User avatar
torontosimpleguy
Posts: 1011
Joined: July 12th, 2004, 5:51 pm

Behaviourial Finance Models

January 15th, 2007, 2:21 pm

QuoteOriginally posted by: yabbadabba[That is a very interesting remark. You suggest, that are two different regimes, one in which the fundamentals are dominant and one in which there is an obvious shift to popularity. This is what Soros described as reflexivity and is quite different to the models in Shleifer's book. These two regimes should have notably different features (volume and momentum). To trade this kind of approach one needs both fundamentals and price information itself. I think technical analysis which is obviously not a scientific theory, aswell as economical theory has surprising little to say about these phenomena. BF might be on the right track, but it has a long way to go. What I'm wondering is which kind of quantitative methods investment or hedge funds use for fundamental equity strategies and for global macro strategies.CFA curriculum contains behavioral finance topic.There are 2 types of traders - rational traders (they trade based on fundamentals) and noise traders (they trade based on "irrationals").Check assigned Reading for LIII behavioral finance topic and you get an idea what is the current state of affairs there.