The Portfolio Selection On A Bayesian Approach

In the 21st century, the financial investment plays an important position in world economic structure. In order to make better investment decisions, the scholars anylaze the theory and practice of portfolios in deeply.Evidences in the developed countries show that the financial portfolio theory for investors plays an important role in financial decision-making.However, empricical studies have shown that asset pricing models have limitations in the late, there are a lot of market irrational reactions in the stock market, such as scale effect, momentum effect, the calendar effect and so on. Compared with foreign mature stock market, the stock market in our courtry also has a lot of irrational behavior, such as the momentum/reversal effect. The typical approaches to portfolio selection essentially reflect two extreme views about the validity of asset pricing models. The first approach regards asset pricing models as useless, and the second approach considers one of these models to be a perfect description of reality.But considersing the reality of market, rather than taking an all-or-nothing approach, one instead might view a particular asset pricing model as a potentially useful (though not completely reliable) tool for financial decision making. This less strict view of the proper role of an asset pricing model is in harmony with the methods of Bayesian decision making.Therefore, this paper tries to make use of the Bayesian methodology to investigate portfolio choices.Many scholars choose the stock to conduct the portfolio, but regarding the momentum in our stock market, this paper attempts to view momentum portfolio and market portfolio as two assets to conduct portfolio. Then, Making use of this Bayesian framework, this paper investigates portfolio choices of hypothetical investors who have varying degrees of confidence in the reliability of asset pricing models, and determines how strongly an investor must believe in a model to justify holding the portfolio the model identifies as optimal.I use a Bayesian methodology developed by Pastor (2000) to examine the portfolio choices of hypothetical mean-variance investors who allocate wealth between the market portfolio and a momentum portfolio..First, Using data for 2000.1 through 2012.12, This paper examines the strength of the momentum effect. Second, the paper examines the return of the momentum portfolio through the CAPM model. Third, Considering these investors have different prior beliefs about the ability of the CAPM to price the momentum portfolio, the paper uses the CAPM theory to form informative prior beliefs in financial decision making.Not surprisingly, investors with less prior confidence in the CAPM eventually allocate more to the momentum portfolio. Somewhat surprisingly, however, an investor with near complete confidence in the CAPM still would have allocated a substantial percentage to the momentum portfolio. This paper examines the strength of the momentum effect from the perspective of asset allocation, and provides investors a new method of financial decision making.