Title: Asymptotic trimming for importance sampling estimators with infinite variance
Authors: Thomas Yang - Australian National University (Australia) [presenting]
Abstract: Importance sampling is a popular Monte Carlo method used in a variety of areas in econometrics. When the variance of the importance sampling estimator is infinite, the central limit theorem does not apply, and estimates tend to be volatile even when the simulation size is large. We consider asymptotic trimming in such a setting. Specifically, we propose a bias-corrected tail-trimmed estimator such that it is consistent and has finite variance. We show that the proposed estimator is asymptotically normal, and has good finite-sample properties in a Monte Carlo study.