Estimating Intertemporal Allocation Parameters using Synthetic Residual Estimation

Abstract

We present a novel structural estimation procedure for models of intertemporal allocation. This is based on modelling expectation errors directly; we refer to it as Synthetic Residual Estimation (SRE). The flexibility of SRE allows us to account for measurement error in consumption and for heterogeneity in discount factors and coefficients of relative risk aversion. An investigation of the small sample properties of the SRE estimator indicates that it dominates GMM estimation of both exact and approximate Euler equations in the case when we have short panels with noisy consumption data. We apply SRE to two panels drawn from the PSID and estimate the joint distribution of the discount factor and the coefficient of relative risk aversion. We reject strongly homogeneity of the discount factors and the coefficient of relative risk aversion. We find that, on average, the more educated are more patient and more risk averse than the less educated. Within education strata, patience and risk aversion are negatively correlated.