The Examine Of The-Day-of-the-Week-Effect And The Analysis Of The Statistical Properties Of The Shanghai Stock Exchange Index

Because the hypothesise about the independent homoscedasticity in the traditional linear regression model is not suitable for describing the price and the income behavior in the financial market, many statisticians and the metrologists start to attempt to describe all kinds of financial market activities well with the improvement methods. Robert ? Engel, an US economist who is the Nobel economic prize winner in 2003 proposed Auto-Regressive Conditional Heteoskedastic (ARCH) model in 1982, Bollerslev proposed Generalized Regressive Conditional Heteoskedastic (GARCH) model in 1986l. Hereafter, some expansive ARCH models were also proposed, formed the group of ARCH models. Because these models can reflect the feature of the financial market well ,they have been widely applied in the time series analysis on financial data.Based on the fundamental theory of multivariate statistical analysis, employing the group of ARCH models, we discussed the non-linear structure of the Shanghai stock exchange index by examining the Shanghai stock exchange index from 1996 to 2005. Finally, we get the best model by studying the standardized residual series with BDS test and comparing the RMSE of the models.On the other hand, with the the-day-of-the-week effect of the Shanghai stock exchange index which were studied by many scholars in recent decades, we use simple multivariate statistical methods, that is investigate whether there is the the-day-of-the-week effect and whether it is stationary by establishing regression model and using Wald test. That is contribute to explore the statistical regularity of the stock market of China and make it standard.