In Bayesian estimation, if bounds on a parameter are specified, will they be applied to truncate the prior, or are they ignored? For example if the following line (taken from Smets and Wouters AER 2007 code) is in the estimated_params; section, will trial values for MCMC be drawn from a normal distribution, or from a truncated normal distribution bounded to [1,3]?
cfc,1.4672,1.0,3,NORMAL_PDF,1.25,0.125;
Thanks,
Michael