make_prior: Construct prior list based on type of sensitivity analysis

In lcomm/rstanmed: Mediation in Stan

Description

Construct prior list based on type of sensitivity analysis

Usage

make_prior(params, prior_type = c("unit", "partial", "strict", "dd",
  "powerprior"), dd_control = NULL)

Arguments

params	List of parameters
prior_type	Variance type: unit variance ("unit"), informative priors on non-identifiable parameters ("partial"), strong prior information ("strict"), data-driven ("dd") based on external or simulated data set, or data pooling/power prior ("pp") based on the external or simulated data set
dd_control	Named list of data-driven options. If external data is to be used, "small_data" needs to be formatted as output from simulate_data. If secondary data source is to be simulated, need a list of parameters as from return_dgp_parameters.

Value

Named list of prior means and variance-covariance matrices

lcomm/rstanmed documentation built on Dec. 6, 2020, 9:11 a.m.