The Comparative Study Between Bayesian Econometrics Modelling And Classic Econometrics Modelling

Bayesian econometrics can be defined as a process that using the parameter in the econometrics as the extraneous variable which has the prior distribution based on the Bayesian theory, then figuring out the posterior distribution according to the Bayes theorem, at last, estimation and hypothesis testing. Bayesian econometrics is based on the rules of probability, which is one of the chief advantages of the Bayesian approach. This can be used any time a researcher is interested in to learn about a phenomenon. In economics we typically work with models which depend upon parameters. Bayesian econometrics is based on a subjective view of probability, which argues that our uncertainty about anything unknown can be expressed using the rules of probability. It has been proved in the period of development that Bayesian theory has plenty of advantage which classical theory don't have. But in the early time, Bayesian theory had been restricted because of the difficulty in the computation. From the twenty century, the computer technology has great development, which made the computation of the posterior distribution based on the multidimensional integral possible. This also promoted the development of the Bayesian Econometrics. Nowadays, more and more people involve in the research of Bayesian EconometricsIn China, there are many people who research Bayesian theory, but people who put the theory into econometrics are not much, so the purpose of my article is to introduce the theory from abroad to China, to make more researchers put their attention on this field. This is significant in the development of Bayesian theoryThis paper put the theory and the practice together, to explain the method of Bayesian analysis and the use of the software, and the steps in building the Bayesian model. At last, use an example to make evidence. In the conclusion, we compare similarities and differences in the model of the two kinds of theory.The conclusion of this paper is following. Firstly, comparing with the frequency, the Bayesian approach makes the full use of the information provided by the sample and the prioror. So in the point estimation, the Bayesian estimation can have better precision. Secondly, in the Bayesian theory, people can say a parameter drops in an interval, but in classical theory, you can't. Thirdly, the Bayesian hypothesis testing, the value of the parameter appears in the way of a ratio, and it allows that the hypotheses are simple.