Exploring the Time-Varying Nature of Beta: An In-Depth Study across S&P 500

The beta coefficient of an asset, or a portfolio, emerges as a fundamental tool to perform the investment risk of the considered asset. However, it is notoriously tough to estimate it, especially considering its time-varying nature. Despite the CAPM assumes that beta remains constant, numerous studies showed its time-varying nature; indeed, a way to overcome the static nature of the CAPM is to analyse beta changing over time. This thesis focuses on the S&P 500 index and its components; it proposes short- and long-run tests through a rolling-windows analysis for beta estimation by applying the OLS estimation to several rolling windows. The final aim is to investigate the time-varying nature of the beta coefficient, how it responds to changes over the different estimation time periods, and to understand the accuracy of the estimated coefficients and the reliability of this approach.