Resampling: an improvement of importance sampling in varying population size models

DOI10.1016/J.TPB.2016.09.002MaRDI QIDQ2399078zbMATH OpenOpenAlexWikidataFDO

Authors C. Merle, R. Leblois, François Rousset, Pierre Pudlo

Publication date 21 August 2017

Published in Theoretical Population Biology (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1603.07237

zbMATH Keywords

coalescentdemographic inferenceimportance samplingjump Markov processpopulation geneticsresampling

Mathematics Subject Classification ID

Sampling theory, sample surveys (62D05) Applications of statistics to biology and medical sciences; meta analysis (62P10) Computational methods for problems pertaining to biology (92-08) General biostatistics (92B15)

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.

Recommendations

• Stopping-Time Resampling for Sequential Monte Carlo Methods
• Inference in Molecular Population Genetics
• Full likelihood inference from the site frequency spectrum based on the optimal tree resolution
• Stepwise mutation likelihood computation by sequential importance sampling in subdivided population models
• Importance sampling on coalescent histories. II: Subdivided population models

Cites work

• Ancestral inference in population genetics
• scientific article; zbMATH DE number 1666094 (Why is no real title available?)
• scientific article; zbMATH DE number 7008406 (Why is no real title available?)
• Importance sampling for the infinite sites model
• Importance sampling on coalescent histories. I
• Importance sampling on coalescent histories. II: Subdivided population models
• Inference in Molecular Population Genetics
• Likelihood methods in statistics
• Monte Carlo strategies in scientific computing.
• Statistical Models
• Stepwise mutation likelihood computation by sequential importance sampling in subdivided population models
• Stepwise mutation model and distribution of allelic frequencies in a finite population.
• The coalescent

Cited in

(7)

• A unified principled framework for resampling based on pseudo-populations: asymptotic theory
• Inferring the demographic history from DNA sequences: an importance sampling approach based on non-homogeneous processes
• Importance sampling for the infinite sites model
• Stopping-time resampling and population genetic inference under coalescent models
• Random weighting-based quantile estimation via importance resampling
• Stopping-Time Resampling for Sequential Monte Carlo Methods
• scientific article; zbMATH DE number 5263157 (Why is no real title available?)

This page was built for publication: Resampling: an improvement of importance sampling in varying population size models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2399078)