Wednesday, May 07, 2014

A thought on decision making

Gary Becker, the Nobel Prize winning economist  died recently. In a tribute in the magazine Business Insider, this remark of his was quoted:
Behavior is driven by a much richer set of values and preferences. The analysis assumes that individuals maximize welfare as they conceive it, whether they be selfish, altruistic, loyal, spiteful, or masochistic. Their behavior is forward-looking, and it is also consistent over time. In particular, they try as best they can to anticipate the uncertain consequences of their actions.
I like the recognition that people make decisions not only on the basis of selfishness and economic gain. I like even more the recognition by Daniel Kahneman, another Nobel Prize winner, that people make choices by different psychological processes at different times.

One possibility is that people seek to make decisions that satisfy a number of different conditions simultaneously. In general there will be many alternatives that will simultaneously satisfy a number of different conditions. Operations researchers sometimes will choose one objective to maximize subject to satisfying the others; I suspect most of us in everyday life are happy to find a solution that is "good enough" when compared to a subset of the most important conditions that occur to us as we make the decision.

I like the use of the term "forward-looking" since it recognizes as a fact that most  of us are not very good at forecasting or prediction.

I also like the recognition that people try to "anticipate the uncertain consequences of their actions".  I would note that there are a couple of different ways. For example, you might seek to anticipate the possibility that your car will break down if you take a long trip. It is more difficult to guess what the cost of such a breakdown might be.

Think about people betting at a (U.S.) race track. In a single race there will be people making pari mutual bets on all the horses. Consider for a second only bets to win. One way to understand the situation is to assume that each punter estimates the probability of each horse winning and the probable pay off if the horse wins; the punter may bet the horse with the best expected return, or may choose not to bet. Of course, the people placing bets at a race track have different amounts of information and different expertise in making bets, but even racing experts with comparable information will come out with different expected results for the horses in a specific race. As they say, that is what makes a horse race.

What that analysis doesn't include is that people like watching horses race, and most people get more enjoyment if they have a bet on the race. So there is an entertainment value in participating quite separate from the expected winnings from a bet. So too, if one is at the races with friends, there is a value in the joint activity of betting the race.

In these circumstances, few if any of us will quantify probabilities and expected returns (or costs). For some, the circumstances will be pathological.

I have tended to think that simulation models exploring the effects of different assumptions of the determinants of human behavior and of their frequency would be a good way to understand "financial" decision making.

No comments: