Off-campus UMass Amherst users: To download dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.
Non-UMass Amherst users, please click the view more button below to purchase a copy of this dissertation from Proquest.
(Some titles may also be available free of charge in our Open Access Dissertation Collection, so please check there first.)
Econometric estimation of risk attitudes
Abstract
This research examines whether the econometric approach, proposed by Antle (1986), can provide reliable estimates of the distribution of risk attitudes in a population. The model measures the risk attitudes of a population which is of interest in policy analysis. Knowing how each input contributes to the moments of output distribution and how decision makers react to those measures enables the policy makers to offer a plan which suits that population. For example in cases where an input such as fertilizer has an increasing effect on the variance of output, a risk averse farmer will use this input less intensively. Therefore, a plan which requires an extensive use of fertilizer may not be accepted by decision makers. It has been hypothesized that there is a systematic relationship between inputs and the first few moments of output distribution. This hypothesis is tested by using a quadratic model for the first three moments of output distribution. The moment functions are found to be heteroskedastic which are corrected by the Glejser technique. The risk attitude distribution estimates are derived by employing the estimated results of the first three moments into a two-stage instrumental variables technique. The empirical results show that: (a) while there is a systematic relationship between the decision variables and the mean and variance of output, there is no significant relationship between the input variables and skewness, which indicates that the output distribution is symmetric; (b) the estimated Arrow-Pratt risk aversion measure indicates that the farmers are moderately risk averse; (c) the degree of dispersion of this measure could not be estimated due to the fact that the estimated variance was negative. The results are compared to experimental estimates of risk attitudes derived by Binswanger. The comparison suggests that this technique can provide an estimate of risk attitude in a population. However, in estimating the degree of dispersion of the estimates this technique is limited.
Subject Area
Agricultural economics|Economics
Recommended Citation
Tabatabai, Nahid, "Econometric estimation of risk attitudes" (1991). Doctoral Dissertations Available from Proquest. AAI9132922.
https://scholarworks.umass.edu/dissertations/AAI9132922