sim_post_normal: Obtain posterior simulations using a normal prior on the...

In carlislerainey/separation: Using informative priors to deal with separation in logistic regression models

Description

sim_post_normal() uses a Metropolis algorithm to obtain posterior simulations using a normal prior on the separating variable and improper, uinform priors on the remaining variables.

Usage

sim_post_normal(formula, data, sep_var, sd = 4.5, n_sims = 1000,
  n_burnin = n_sims/2, n_thin = 1, n_chains = 3, n_cores = n_chains,
  tune = 1)

Arguments

formula	A logistic regression model.
data	A data frame.
sep_var	The name of the separating variable.
sd	The standard deviation of the normal prior on the coefficient for the separating variable.
n_sims	The number of simulations after the burn-in period.
n_burnin	The number of burn-in iterations for the sample.
n_thin	The thinning interval used in the simulation. The number of MCMC iterations must be divisible by this value.
n_chains	The number of MCMC chains being run.
n_cores	The number of MCMC cores. Defaults to the number of chains.
tune	The tuning parameter for the Metropolis sampling. Can be either a positive scalar or a (k+1)-vector, where k is the number of variables in the model. Presently passed to MCMCmetrop1R.

References

Rainey, Carlisle. "Dealing with Separation in Logistic Regression Model." Working paper. Available at http://crain.co/papers/separation.pdf.

carlislerainey/separation documentation built on May 13, 2019, 12:45 p.m.