The coefficient of relative risk aversion: a Monte Carlo study investigating small sample estimator problems

When estimating consumer Euler equations using the Generalized Method of Moments, negative estimates for the coefficient of relative risk aversion are frequently found. In this paper we suggest that this anomaly may be due to a small sample bias. Based on Monte Carlo simulations we show that the estimate of the coefficient of relative risk aversion tends to have a negative bias. Our results suggest that the other estimates in the equation are not (as seriously) biased. Inference is problematic, however, since the standard errors of the estimates tend to be underestimated. We illustrate these points through the estimation of an Euler equation using annual data for Belgium.