Efficient Estimation of Risk Attitude with Seminonparametric Risk Modeling

Recent development in production risk analyses has raised questions on the conventional approaches to estimating risk preferences. This study proposes to identify the risk separately from input equations with a seminonparametric estimator. The approach circumvents the issue of arbitrary risk specifications. Meanwhile, it facilitates analytical derivation of input equations. The GMM estimation method is then applied to input equations to estimate risk preferences. The procedure is validated by a Monte Carlo experiment. Simulation results show that the proposed method provides a consistent estimator and significantly improves estimation efficiency.