Does the day-of-the-week effect on volatility improve the volatility forecasts?

Abstract

This study tests the presence of the day-of-the-week effect on stock market volatility of six European stock markets. Using a GARCH or GARCH-GJR specifications for the variance equation, we find that the day of week effect is present in volatility equation. Finally, we test whether the statistically significant in-sample findings regarding seasonality in volatility lead to better out-of-sample forecasts of volatility.

Notes

¹ Results are not reported to save space but they are available from the author upon request.

² An alternative way to model seasonality in stock returns is the periodic autoregressive model with periodic integrated GARCH process proposed by Franses and Paap (2000) and the periodic stochastic volatility process developed by Tsiakas (2006).

³ The model 1 is the ARMA(p,q)-GJR-GARCH model. Results are not reported to save space but they are available from the author upon request. The GARCH-GJR model has been introduced by Glosten et al. (1996).

⁴ Results are available from the author upon request.

Table 3. Estimates of model 2