param_prop_setup — param_prop

Set up proposed new log parameter values for next iteration in chain

Usage

param_prop_setup(log_params = c(), chain_cov = 1, adapt = 0)

Arguments

log_params: Previous log parameter values used as input
chain_cov: Covariance calculated from previous iterations in chain
adapt: 0/1 flag indicating which type of covariance to use for proposition value (TBA) '

Details

Takes in current values of parameter set used for Markov Chain Monte Carlo fitting and proposes new values from multivariate normal distribution where the existing values form the mean and the standard deviation is based on the chain covariance or (if the flag "adapt" is set to 1) a flat value based on the number of parameters.