https://github.com/cran/bayestestR
Revision 85d0a0463621c30063752c5577e591be0de5ae60 authored by Dominique Makowski on 26 July 2021, 08:40:08 UTC, committed by cran-robot on 26 July 2021, 08:40:08 UTC
1 parent 601edcd
Raw File
Tip revision: 85d0a0463621c30063752c5577e591be0de5ae60 authored by Dominique Makowski on 26 July 2021, 08:40:08 UTC
version 0.10.5
Tip revision: 85d0a04
check_prior.Rd
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/check_prior.R
\name{check_prior}
\alias{check_prior}
\title{Check if Prior is Informative}
\usage{
check_prior(model, method = "gelman", simulate_priors = TRUE, ...)
}
\arguments{
\item{model}{A \code{stanreg}, \code{stanfit}, or \code{brmsfit} object.}

\item{method}{Can be \code{"gelman"} or \code{"lakeland"}. For the
\code{"gelman"} method, if the SD of the posterior is more than 0.1 times
the SD of the prior, then the prior is considered as informative. For the
\code{"lakeland"} method, the prior is considered as informative if the
posterior falls within the 95\% HDI of the prior.}

\item{simulate_priors}{Should prior distributions be simulated using
\code{\link{simulate_prior}} (default; faster) or sampled via
\code{\link{unupdate}} (slower, more accurate).}

\item{...}{Currently not used.}
}
\value{
A data frame with two columns: The parameter names and the quality
  of the prior (which might be \code{"informative"}, \code{"uninformative"})
  or \code{"not determinable"} if the prior distribution could not be
  determined).
}
\description{
Performs a simple test to check whether the prior is informative to the posterior. This idea, and the accompanying heuristics, were discussed in \href{https://statmodeling.stat.columbia.edu/2019/08/10/}{this blogpost}.
}
\examples{
\dontrun{
library(bayestestR)
if (require("rstanarm")) {
  model <- stan_glm(mpg ~ wt + am, data = mtcars, chains = 1, refresh = 0)
  check_prior(model, method = "gelman")
  check_prior(model, method = "lakeland")

  # An extreme example where both methods diverge:
  model <- stan_glm(mpg ~ wt,
    data = mtcars[1:3, ],
    prior = normal(-3.3, 1, FALSE),
    prior_intercept = normal(0, 1000, FALSE),
    refresh = 0
  )
  check_prior(model, method = "gelman")
  check_prior(model, method = "lakeland")
  plot(si(model)) # can provide visual confirmation to the Lakeland method
}
}
}
\references{
https://statmodeling.stat.columbia.edu/2019/08/10/
}
back to top