An Empirical Study On Turn-of-the-month Effect And Its Window Dress Hypothesis In Chinese Stock Market

According to weak-form efficient market hypothesis theory, all historical transaction information will be reflected in asset prices quickly and fully, and the investors can not profit by predicting the future returns from the model and characteristics of past returns. In other words, return on assets does not appear to changes in the characteristics of regularity, and technical analysis by analyzing historical transaction information of stocks is invalid.However, there have been a large number of empirical studies which have shown a various anomalies of stock market since the eighties of the last century. Anomaly is the phenomenon which can not be explained by Capital Asset Pricing Model, for example, the risk-adjusted rate of returns of some characteristics of securities is still higher than the expected rate of return. The Turn of Month Effect is one of anomalies of stock market. It is the return of securities will be much higher than other days during the last day of previous month to the first three days of the next month. The existence of the Turn-of-the-Month Effect constitutes a strong challenge to weak-form efficient market hypothesis.In order to demonstrating the possibility of the Turn of month effect, we will study the defects of EMH firstly. Then, we will make an empirical test about the stock Indexes of Shanghai and Shenzhen, by the model of ARMA-GARCH, to test the existence of the Turn of Month Effect. In this paper, we discover that the turn of month effect widespread exists in Shanghai and Shenzhen stock markets, and so it is in rolling sample tests. To find the reason of the turn of month effect in Shanghai and Shenzhen stock markets, we construct a hypothesis test according to the characteristics of the development of China's fund industry, and we find that, the turn of month effect increases markedly after the advent of Securities Investment Fund (1998-2004), while it weakens in some degree in the period of fund industry which is relatively standardized (2005-2009). It shows that the window dress hypothesis has a considerable explanatory power on the turn of month effect of China's stock market.The existence of turn of month effect means defects in industry supervision and information disclosure of China's stock market, so we will try to give some policy advices at the end of this paper.