COMPARISON BETWEEN WAVELET BAYESIAN AND BAYESIAN ESTIMATORSTOREMEDYCONTAMINATION IN LINEAR REGRESSION MODEL

Abstract

This paper proposes a new method that depends on the wavelet shrinkage and linking it to a Bayesian estimate and it is called wavelet Bayesian method for reducing the effect of contamination and then relying on the outputs of this method to estimate the multiple regression modelthrough using simulation experiment for several of data contamination and real data. The comparing results between the proposed method with traditional Bayesian methodbased on the statistical criterion(RMSE).It was concluded that the Bayesian estimation using wavelet shirking filter gives the best results and more accurate thantraditional method for all simulations and real data based on RMSE criterion.