# Estimating risk parameters from observed choices [migrated]

My background is in labor economics/applied micro but I am currently working with data collected from a field experiment. My experimental/micro skills are rusty to say the least.

In my data, I observe individuals choosing one of the choices $A, B, C, D$. The value of the choice is realized after the choice. I know that

$$E(A) <= E(B) <= E(C) <= E(D)$$ $$VAR(A) <= VAR(B) <= VAR(C) <= VAR(D)$$

If I assume that $A, B, C, D$ are continuous, independent and normally distributed and individuals are aware of expected payoffs $E(i)$ and variances $VAR(i)$.

• Is there a way to estimate a expected risk aversion parameter (e.g. CRRA) for a group of people?
• Would it make sense to compare these risk estimates from two groups of people, say men and women or high educated and low educated?

I would appreciate if you guys could come up with some relevant literature.

-

## migrated to stats.stackexchange.com by Turukawa♦May 2 '12 at 8:04

This question belongs on our site for people interested in statistics, machine learning, data analysis, data mining, and data visualization.

I quess I should add, that outcomes from lotteries are by construction continuous but nothing limits them to be positive. –  o.k. Feb 14 '12 at 16:02
Are these lotteries for money? You'd need to control for diminishing marginal utility of money before you could elicit risk attitudes, no? I've read stuff on Prospect theory which discusses how to elicit these sorts of attitudes. –  Seamus Feb 14 '12 at 23:27
Thanks for your reply. I am aware of prospect theory and I would probably be able to control for wealth. This goes somewhat beyond my original question, though. I would like to find references on estimating preference parameteres 1.) from field experiment data 2.) when outcomes are continuous instead of discrete. I have not yet encountered an experimental paper with other than discrete lotteries. –  o.k. Feb 15 '12 at 15:58
jstor.org/stable/10.2307/30034997 This paper is a good start. –  Dimitris Feb 16 '12 at 15:31