Modelling the Stock Market Volatility of Dar es Salaam Stock Exchange (DSE) using Generalized Autoregressive Conditional Heteroscedasticity

Abstract

The existing empirical literature has extensively explored stock market return volatility in various emerging and developing markets; however, limited attention has been given to the Dar es Salaam Stock Exchange (DSE). This study seeks to address this gap by analyzing the volatility dynamics of stock returns in the DSE. The analysis is based on a dataset comprising 1,846 daily observations spanning the period from June 2014 to November 2021. Consistent with prior studies, the findings reveal a significant negative relationship between returns and risk, as modeled using the AR(1)-GARCH(1,1)-M framework. The application of the GARCH(1,1) model effectively captures volatility clustering, following the confirmation of heteroscedasticity in the return series. However, due to the GARCH model’s limitations in capturing asymmetries in volatility (i.e., the leverage effect), the analysis was extended using the AR(1)-EGARCH model. The results support the presence of a leverage effect in the DSE, indicated by a negative and statistically significant leverage coefficient. This suggests that negative shocks have a greater impact on volatility than positive shocks of the same magnitude. Moreover, the study confirms a negative correlation between stock returns and volatility. These findings imply that higher levels of risk may lead to disproportionately larger losses for investors in the DSE. Therefore, market participants, policymakers, and portfolio managers must exercise caution and implement robust risk management strategies to safeguard investments against unexpected market fluctuations. The results also offer valuable insights for investors, scholars, and researchers interested in understanding the behavior of stock return volatility in frontier markets such as Tanzania.