Abstract
Sampling/importance resampling method is widely used in Bayesian statistics. Literatures have carried out a variety of improvements to promote the computational expanse and the accuracy. In this work, further improvement based on quasi-random sampling/importance resampling method is derived. The randomized version of quasi-random sampling/importance resampling algorithm is proposed, and the performance is illustrated through three examples. The empirical study shows that the proposed approaches are more accurate than sampling/importance resampling method and quasi-random sampling/importance resampling method. Moreover, it can be used to estimate the Monte Carlo error, while the quasi-random sampling/importance resampling method has constraints in this problem.
Acknowledgments
We thank the anonymous referee for his/her helpful suggestions. The second author is partially supported by a grant from the National Natural Science Foundation of China (No. 11571133 and 11471135).
Correction Statement
This article has been republished with minor changes. These changes do not impact the academic content of the article.