Generate initial values for MCMC from the prior

Generates initial values for Markov Chain Monte Carlo (MCMC) by sampling from the prior distributions of the model parameters. This helps in starting the MCMC chains from different points in the parameter space.

Usage

prior_inits(dat, chains, model = "drm")

Arguments

dat

A list containing the data and prior parameters, typically generated by make_data() or make_data_sdm().

chains

An integer specifying the number of MCMC chains to initialize. Defaults to 4.

model

A character string specifying the model type. Must be either "drm" or "sdm". Defaults to "drm".

Value

A list of lists, where each inner list contains initial values for one MCMC chain. The structure of each inner list is the same as the output of prior_sample().

See also

prior_sample()

Author

lcgodoy