Bayesian Analysis of Time-Varying Systematic Risk: Evidence from the Warsaw Stock Exchange
Abstract
This study investigates the time variation of systematic risk for companies listed on the Warsaw Stock Exchange (WSE) during the period 2011 – 2025. The empirical analysis combines rolling-window estimation with Bayesian approaches to structural instability and stochastic risk dynamics. The results reveal instability of beta coefficients, particularly during the COVID-19 pandemic and the subsequent energy and inflation shocks. The evidence for structural changes is considerably stronger for residual variance than for beta coefficients themselves. Dynamic Bayesian models further indicate that systematic risk evolves continuously over time and exhibits persistent but stationary dynamics, strongly rejecting random-walk-type behavior of beta coefficients. Banking sector companies display relatively stable market exposure, whereas growth-oriented and energy-related firms exhibit significantly stronger beta instability. The findings suggest that the assumption of constant beta coefficients may be difficult to justify for companies listed on the WSE and support the application of dynamic asset pricing models in emerging equity markets.