Winsorized Importance Sampling

Winsorized Importance Sampling

Abstract

This talk considers making the usual importance sampling estimator more robust via winsorization. The threshold level at which to winsorize is determined by a concrete version of the Balancing Principle, which may be of independent interest. The procedure adaptively chooses a threshold level among a pre-defined set by roughly balancing the bias and variance of the estimator when winsorized at different levels. As a consequence, it provides a principled way to perform winsorization, with finite-sample optimality guarantees. Empirically, the estimator outperforms the usual importance sampling estimator in high-variance settings, and matches the performance of the usual estimator when the variance of the importance sampling weights is low.

Date
Event
Location
Stanford, USA
Avatar
Paulo Orenstein
PhD candidate in Statistics