| Mutual fund performance evaluation has been one of the important fields that foreign scholars in finance keep on researching. In recent years, we pay more and more attention to this problem of fund performance evaluation with the rapid development of domestic securities investment funds. Nevertheless, our research into fund performance evaluation is relatively backward because most of the available researches are confined into the mechanical application of traditional risk-adjusted indices with few new methods being proposed or introduced. In the meanwhile, these methods are usually based on historical data and thus disregard the practical procedure, the combination of intuition and data analysis, by which investors make their investment decision. For this reason, the thesis specially introduces the theory of Bayesian Statistical Inference into performance evaluation procedure for securities investment funds and proposes a Bayesian method of performance evaluation from an investor' perspective. Our propose is to provide a new lens on the performance evaluation evidence and enriches the theory of performance evaluation for securities investment funds. We begin with a flexible set of prior beliefs about managerial skill that can be elicited without any reference to probability distribution or their parameters. We then combine these prior beliefs with a general multi-factor model and derive an analytical solution for the posterior expectation of "alpha", the intercept term from the model. The solution can be computed using only a few extra steps beyond maximum likelihood estimation. Finally, we apply our methodology to a sample of domestic securities investment funds and make practical analysis. Overall, the Bayesian method not only shows us how prior beliefs influence investors' final decision, but also relatively solves the problem that the general frequentist results are to the most extent influenced by the factors of data precision and significance level.
|
PDF Full Text Request