Sampling with replacement vs Poisson sampling: a comparative study in optimal subsampling

TL;DR

This paper compares subsampling with replacement and Poisson sampling, analyzing their estimation efficiency and providing optimal probabilities to improve subsampling methods for large data sets.

Contribution

It offers a rigorous theoretical comparison of the two sampling procedures and derives optimal subsampling probabilities for improved estimation efficiency.

Findings

Poisson subsampling may have higher estimation efficiency.

Optimal subsampling probabilities minimize estimator variance.

Algorithms based on optimal probabilities are effective in practice.

Abstract

Faced with massive data, subsampling is a commonly used technique to improve computational efficiency, and using nonuniform subsampling probabilities is an effective approach to improve estimation efficiency. For computational efficiency, subsampling is often implemented with replacement or through Poisson subsampling. However, no rigorous investigation has been performed to study the difference between the two subsampling procedures such as their estimation efficiency and computational convenience. This paper performs a comparative study on these two different sampling procedures. In the context of maximizing a general target function, we first derive asymptotic distributions for estimators obtained from the two sampling procedures. The results show that the Poisson subsampling may have a higher estimation efficiency. Based on the asymptotic distributions for both subsampling with replacement and Poisson subsampling, we derive optimal subsampling probabilities that minimize the variance functions of the subsampling estimators. These subsampling probabilities further reveal the similarities and differences between subsampling with replacement and Poisson subsampling. The theoretical characterizations and comparisons on the two subsampling procedures provide guidance to select a more appropriate subsampling approach in practice. Furthermore, practically implementable algorithms are proposed based on the optimal structural results, which are evaluated through both theoretical and empirical analyses.