To read this content please select one of the options below:

Robust Estimation and Inference for Importance Sampling Estimators with Infinite Variance*

Joshua C. C. Chan

Chenghan Hou

Thomas Tao Yang

^aPurdue University, USA

^bHunan University, China

^cAustralian National University, Australia

Essays in Honor of Cheng Hsiao

ISBN: 978-1-78973-958-9, eISBN: 978-1-78973-957-2

Publication date: 15 April 2020

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 erratic even when the simulation size is large. The authors consider asymptotic trimming in such a setting. Specifically, the authors propose a bias-corrected tail-trimmed estimator such that it is consistent and has finite variance. The authors show that the proposed estimator is asymptotically normal, and has good finite-sample properties in a Monte Carlo study.

Keywords

Acknowledgements

Acknowledgments

The authors thank Tong Li, a referee, and participants of the 2018 AIE conference in honor of Cheng Hsiao for helpful comments.

Citation

Chan, J.C.C., Hou, C. and Yang, T.T. (2020), "Robust Estimation and Inference for Importance Sampling Estimators with Infinite Variance*", Li, T., Pesaran, M.H. and Terrell, D. (Ed.) Essays in Honor of Cheng Hsiao (Advances in Econometrics, Vol. 41), Emerald Publishing Limited, Leeds, pp. 255-285. https://doi.org/10.1108/S0731-905320200000041008

Publisher

:

Emerald Publishing Limited