Overreaction, Heteroscedasticity, and Spillovers in Stock Returns: Evidence From the Kuwait Stock Exchange

This research examines stocks\u27 returns and volatility in the Kuwait Stock Exchange (KSE). The research is organized in five chapters. The first chapter provides an introduction of the research, its importance, and its main goals. The second chapter presents a historical background of the KSE and the stages it experienced to reach its current situation. The main goal of this chapter is to pave the road for the subsequent chapters. Then, the third chapter tests the overreaction hypothesis using monthly data for stocks listed on the Kuwait Stock Exchange over the 1993–2002 period. Similar to the findings of De Bondt and Thaler (1985, 1987), this chapter finds significant systematic price reversals for stocks that experience extreme long-term gains or losses: Past losers significantly outperform past winners. This is consistent with the behavioral hypothesis of investor overreaction. The results indicate that, on average, abnormal returns in January are not the main driver behind the overreaction phenomenon in the KSE. The fourth chapter investigates the stock index returns volatility in the KSE for the period from 1984 through 2003. Models of autoregressive conditional heteroscedasticity (ARCH), generalized autoregressive conditional heteroscedasticity (GARCH), and GARCH-in-Mean (GARCH-M) are fitted to the index returns. The results strongly indicate that the returns series exhibit significant ARCH and GARCH effects with non-normality. A GARCH(1,1) model is demonstrated to be a suitable specification of the process. Additionally, the findings show that the persistence of volatility shocks is substantial. Finally, chapter five examines return and volatility spillover effects among the eight sectors listed in the KSE for the period of 1992 through 2003. This chapter also analyzes the impact of world oil price fluctuations as well as the day of the week effect on sectors\u27 returns and volatility. Using GARCH and TARCH models, the findings indicate statistically significant spillover effects from the Food sector to the rest. These spillover effects are negative on sectors\u27 returns, but positive on sectors\u27 volatilities. The results also demonstrate that movements in world oil prices have no significant influence on sector\u27s returns or volatilities in the KSE. In addition, the day of the week effect is found to be present in sectors\u27 returns as well as in sectors\u27 volatilities. Finally, the estimation results of the TARCH(1,1) model indicate no statistically significant asymmetric information effects