evidenceRatio: Obtain evidence ratios comparing two models

In abnormally-distributed/Bayezilla: Bayesian Models With JAGS and Several Utilities for Data Exploration and Model Evaluation.

Description

This function is used to compare two models. The output can be interpreted model compared to the other. Instead of using marginal likelihoods this function utilizes either the Watanabe-Akaike Information Criterion or the Leave-One-Out Cross-Validation Information Criterion. The default is to use LOO-IC. See the IC() function for more information on how these are calculated. Models with a larger evidence ratio are favored over the alternative. Posterior model probabilities are also provided for each model to aid interpretation.

Usage

evidenceRatio(modelA, modelB, method = "LOO")

Arguments

modelA	the first model
modelB	the second model
method	one of "LOO" (the default) or "WAIC"

Value

a data frame

Examples

evidenceRatio()

abnormally-distributed/Bayezilla documentation built on Oct. 31, 2019, 1:57 a.m.