rhats: A Gelman-Rubin convergence diagnostic function.

In tbrown122387/mmcmc: Miscellaneous Markov Chain Monte Carlo Functions.

Description

This function calculates the Gelman-Rubin convergence diagnostic (Rhat) on some samples. It assumes each chain is in a separate csv file, and each file contains the samples for all parameters. For each chain, this function splits each column into two, which means "m" comes out to be twice the number of csv files. Finally, this function assumes each csv has a one-line header.

Usage

rhats(chain_files, burn = 0, ...)

Arguments

chain_files	vector of string paths for each file of samples
burn	the number of rows you want to discard as burn in. Defaults to 0.
...	arguments to be passed in to lapply

Examples

files <- c('/fake/path/samps1.csv', '/fake/path/samps2.csv')
rhats(files, burn = 300)

tbrown122387/mmcmc documentation built on Dec. 31, 2019, 12:34 p.m.