The Day-of-the-week Effect In The Individual Shares Of China’s Insurance Industry

The potential definition of day-of-the-week effect is that a particular trading day’sstock returns on average are less than stock returns on other days of the week and theaverage stock returns are statistically significantly positive or negative. The first paperabout it was published and found that stock daily returns are significantly negative onMondays and positive on Fridays in the U.S. stock market, so it is also called Mondayeffect or weekend effect.After reviewing relative literature, we find that there is lots of literature examiningthe calendar anomalies in stock indexes, but the papers studying anomalous stock returnsin the individual shares of China’s insurance industry are few. In addition, for the researchmethods, most existing papers like choosing to use traditional ways such as the linearregression analysis, the GARCH series model, linear quantile regression model and so on,whereas no-normal and quite heavy tails characteristics of the stock return disturbancemakes the analysis based on these methods to be challenged.Therefore, the purpose of this paper is to investigate empirically the existence of theday-of-the-week effect in individual shares of China’s insurance industry and comparewhether insurance companies listed in different places will have affects on themselvesusing stochastic dominance (SD) criterion. In this study, we use two SD tests, one isproposed by Davidson and Duclos (DD test) and the other is put forward by Barrett andDonald(Consistent test). Besides, we have applied Monte Carlo simulation to compare thesize and power of these two tests.After analysis, we have the following conclusion. First, our traditional testingapproaches show that the lowest and significantly negative mean daily return is observedon Tuesday and the highest volatility occurs on Monday in individual shares of China’sinsurance industry. Second, our SD tests show that there is no evidence of Monday effectbut Tuesday effect in the weak forms in some individual shares of China’s insuranceindustry, which contrasts to those stock markets in developed countries. Morever, thereturn’s dominant relation of most trading days can be judged at the second-order. Third, our Monte Carlo simulation results show that the heavy tails characteristics oft-distribution have no effects on the size and power of the consistent test, in particular, theconsistent test is superior to the DD test in terms of empirical size, and performs as well asthe DD test in terms of empirical power. Finally, both the two methods show that thereexists the evidence of Tuesday effect in individual shares of China’s insurance industryand the choice of listed places has no effects on themselves.