posterior.predictive.sample: Generates posterior predictive samples

In BEDASSLE: Quantifies Effects of Geo/Eco Distance on Genetic Differentiation

Generates posterior predictive samples

Description

This function simulates data using the inference model parameterized from the joint posterior of the MCMC and the observed independent variables (D_{ij} and E_{ij}). These posterior predictive samples can be compared to the observed data to see how well the model is able to describe the observed data.

Usage

posterior.predictive.sample(MCMC.output, posterior.predictive.sample.size, output.file, 
prefix = "")

Arguments

MCMC.output	The standard MCMC output file generated from a BEDASSLE run.
posterior.predictive.sample.size	The number of posterior predictive datasets the user wishes to simulate.
output.file	The name that will be assigned to the R object containing the posterior predictive datasets. The suffix ".Robj" will be appended to the user-specified name.
prefix	If specified, this prefix will be added to the output file name.

Details

This function simulates datasets like those the user analyzed with BEDASSLE, using the same independent variables (sample.sizes, D_{ij} and E_{ij}) as in the user's dataset and plugging them into the inference model, which is parameterized by randomly drawing parameter values from the joint posterior output of the MCMC analysis. These posterior predictive simulated allelic count data are summarized as unbiased pairwise F_{ST} (using calculate.all.pairwise.Fst), which may then be compared to the observed unbiased pairwise F_{ST} to determine how well the model is able to describe the user's data. The output of posterior.predictive.sample can be visualized using plot.posterior.predictive.sample.

Author(s)

Gideon Bradburd

BEDASSLE documentation built on April 11, 2022, 1:07 a.m.