savage.dickey.bf: Savage-Dickey density ratio

In mattelisi/mlisi: Miscellaneous R helpers and functions

Description

Approximate the Savage-Dickey density ratio, and compute Bayes factor in favour of null hypothesis. As input it takes a vector of samples, which should represent the posterior distribution of the quantity of interests, usually obtained through MCMC sampling. The function assumes a Gaussian prior (the default is mean 0 and sd 1).

Usage

savage.dickey.bf(x, x_0 = 0, prior.mean = 0, prior.sd = 1,
  plot = F)

Arguments

x	vector of samples from posterior distribution
x_0	location of point-null hypothesis (usally zero)
prior.mean	mean of Gaussian prior used to estiamte the model
prior.sd	standard deviation of Gaussian prior used to estiamte the model
plot	logical: should the prior and posterior probability be plotted?

Value

Bayes factor, indicating how likely are the data under the null hypothesis, compared to the alternative.

mattelisi/mlisi documentation built on Oct. 13, 2019, 5:59 p.m.