| Summary: | This paper analyzes the day-of-the-week effect in 9 stock indices, between 1992-2012, to determine whether there is a relation between this calendar anomaly and the level of market development (using the MSCI Market Classification list: that divides the markets by: Developed, Emerging and Frontier). Moreover, we also analyze if the significance and the estimates for the coefficient of the daily effect are influenced by the international crisis – Asian crisis and Subprime crisis, by using 5 sub-periods: 1992-1996 (pre-Asian crisis), 1997-1998 (Asian crisis), 1999-2003 (post-Asian crisis); 2004-2006 (pre-Subprime crisis) and 2007-2012 (Subprime crisis). In order to do this, we use the OLS regression model (applied in previous studies) and the TGARCH to model the error’s conditional variance. We conclude that the day-of-the-week effect is not related with the level of market development, but find evidence to assert that the international crisis have an impact on the significance and the estimates for the coefficients of the daily effect. We also corroborate that the analysis of the day-of-the-week effect in long period samples may not be the best method to understand the real behavior of this anomaly.
|
|---|