Resampling: an improvement of Importance Sampling in varying population size models

TL;DR

This paper introduces a resampling technique to improve the efficiency of sequential importance sampling algorithms for likelihood estimation in models with varying population sizes, significantly reducing computational costs.

Contribution

It develops a novel resampling method for importance sampling that enhances likelihood inference in demographic models with changing population sizes.

Findings

Reduced computational cost by up to 100 times in some cases

Improved likelihood estimation accuracy in variable population size models

Extended applicability of importance sampling to more complex demographic scenarios

Abstract

Sequential importance sampling algorithms have been defined to estimate likelihoods in models of ancestral population processes. However, these algorithms are based on features of the models with constant population size, and become inefficient when the population size varies in time, making likelihood-based inferences difficult in many demographic situations. In this work, we modify a previous sequential importance sampling algorithm to improve the efficiency of the likelihood estimation. Our procedure is still based on features of the model with constant size, but uses a resampling technique with a new resampling probability distribution depending on the pairwise composite likelihood. We tested our algorithm, called sequential importance sampling with resampling (SISR) on simulated data sets under different demographic cases. In most cases, we divided the computational cost by two for the same accuracy of inference, in some cases even by one hundred. This study provides the first assessment of the impact of such resampling techniques on parameter inference using sequential importance sampling, and extends the range of situations where likelihood inferences can be easily performed.