Volatility Modeling and Forecasting in Stock Markets: A Comparative analysis

Abstract

This study compares ARCH, GARCH, stochastic volatility models, as well as machine learning-based methods as it explores the crucial field of volatility modeling and forecasting in stock markets. Using a descriptive design in addition to secondary data collection, the study adopts a deductive approach and interpretivist philosophy. The results demonstrate the GARCH model's strong ability to predict volatility, particularly in times of increased market turbulence. Different results are obtained using machine learning-based methods and stochastic volatility models, highlighting the importance of careful model selection. The sensitivity analysis provides useful information for practitioners by highlighting the GARCH model's robustness to parameter variations. The present study offers significant insights into the dynamic field of financial market analysis, providing direction for future research endeavors in addition to risk mitigation tactics.