Investor Sentiment and Stock Market Volatility in the Egyptian Exchange: A GARCH and EGARCH Analysis with Persistence and Asymmetry Checks (2009-2025)

Abstract

This study empirically investigates the impact of investor sentiment on stock market volatility in the Egyptian Stock Exchange (EGX) over the period 2009–2025, using daily data from 22 non-financial firms listed on the EGX 30. A composite sentiment index (SMI) is constructed using Principal Component Analysis (PCA) as the primary aggregation method, with a simple average approach used for robustness, based on four market-based proxies: stock turnover ratio (STURN), money flow index (MFI), advancing-to-declining ratio (ADR), and relative strength index (RSI). We employ a GARCH(1,1) model with SMI in the conditional variance equation to capture volatility clustering and persistence, and we extend the analysis to an EGARCH(1,1) specification to test for asymmetric effects of positive versus negative sentiment. The transmuted normal distribution is implemented in the main empirical analysis to model the error term distribution more flexibly, alongside quasi-maximum likelihood estimation with robust standard errors. We also incorporate macroeconomic controls (inflation, interest rates, GDP growth, political instability) and conduct Granger causality tests to address endogeneity concerns. Diagnostic tests confirm the absence of multicollinearity and remaining heteroscedasticity. The results show that investor sentiment has a positive and statistically significant impact on volatility (coefficient =0.278, p < 0.001). The GARCH(1,1) results reveal high volatility persistence (α + β = 0.97), supporting the hypothesis that sentiment shocks have long-lasting effects. The EGARCH model indicates asymmetry: negative sentiment shocks increase volatility more than positive shocks of the same magnitude (asymmetry coefficient γ = −0.073, p < 0.05). Granger causality tests provide evidence of bidirectional causality between sentiment and volatility. Out-of-sample forecasting demonstrates that the SMI has modest but significant predictive power for volatility, though the results should be interpreted as suggestive rather than definitive. These findings contribute to behavioral finance literature by providing evidence from an under-researched frontier market (Egypt) and offer potential tools for regulators and investors to monitor sentiment as an early-warning indicator of excessive volatility.