The anaysis pricing of the diamonds with Bayesian statistics

The aim of this study is to analyse the factors which determine the levels of the prices of diamonds by a Bayesian regression analysis. Some parameter estimation techniques are also mentioned in this study. The importance of the maximum likelihood method in parameter estimation is emphasized and the relation between the likelihood function and Bayesian approach is put forward. Besides some concepts of prior and posterior distributions are discussed to some extent. Some of the affects of prior distributions on posterior distributions are exposed. On the empirical part of this study, a regression model is constructed by using the factors which determine the price levels of diamonds. By unifying the information obtained from prior assumptions and the data at hand a new regression model is developed. Thus it is observed that the lengths of the confidence intervals for the parameters have been reduced. Key Words: prior distribution , posterior distribution , bayesian analysis