BAYESIAN ESTIMATION AND AN APPLICATION OF GIBBS SAMPLING TECHNIQUE AND RANDOM WALK METROPOLIS - HASTING ALGORITHM IN TWO PHASE LINEAR REGRESSION MODEL

Abstract

Here, in this research paper, we have applied the Gibbs Sampling Technique and RWM-H (Random Walk Metropolis - Hasting) Algorithm for the Bayesian Estimation of m, β1, β2 and 1/2. Also we have assumed that at some point of time say 'm', the co-efficient of regression changes from β1 to β2. Further, we have discussed about the effects of prior information on the Bayes estimates on the basis of the TPLR (Two Phase Linear Regression) Model with a Bayesian approach.