Unconditional Estimation of Time-Varying-Parameter Models: A Gibbs-Sampling Approach
Chung-ki Min · RePEc: Research Papers in Economics · 1997
This study addresses the issue of unconditional estimation of regression models with time-varying parameters. Using a data augmentation in which unobserved random coefficients are treated as missing data, procedures for the Gibbs sampler are developd. Several examples are presented to illustrate how the Gibbs-sampling procdedures perform in practice.