An empirical study of beta-coefficient in Shanghai Stock Market: A non-parametric approach

Beta-coefficient, the key parameter of CAPM model, is widely adopted in theoretical research and financial practices such as asset pricing, portfolio management, and fund performance evaluation. At present, accuracy in the measurement of the beta coefficient and employment & improvement of estimation approach are striking topics. Since 1990s, the commonly-used beta-coefficient estimation models have been employed. No new model or method has emerged since the DCC-MV-GARCH model firstly used by Engle (2002), until Kauermann and Semmler (2007) addressed a nonparametric method based on the generalized additive model. The model has not been applied to study beta of China Stock Market. Thus, this paper firstly used the generalized additive model for an empirical study of shanghai stock data and resulted in more reliable estimate of beta as well as some explanation of stock market phenomenon.