Dynamic Connectedness of Eastern European Stock Markets: An Extended Joint Connectedness Approach

Abstract

This paper examines the dynamic return spillovers among ten Eastern European stock markets using an extended joint connectedness approach. We analyze daily log returns from 2010 to 2024 for equity indices of Serbia, Bosnia and Herzegovina, Croatia, Slovenia, North Macedonia, Romania, the Czech Republic, Hungary, Poland, and Ukraine. We employ the Diebold-Yilmaz connectedness framework, augmented with a Time-Varying Parameter VAR (TVP-VAR) model to capture evolving relationships, and incorporate the extended joint spillover methodology to address bias from normalization. The Connectedness Approach R package is used for implementation. Our results show that these markets exhibit a moderate degree of interconnectedness on average - the Total Connectedness Index is around 25%, indicating that roughly one-quarter of forecast variance is due to cross-market shocks. Spillovers are highly time-varying, with pronounced surges during major crises such as the European sovereign debt crisis, the COVID-19 pandemic, and the Russia-Ukraine conflict. Directional spillover analysis reveals that larger markets (e.g., Romania, Poland, Czech Republic) tend to be net transmitters of shocks, whereas smaller frontier markets (e.g., Serbia, Bosnia, North Macedonia) are net recipients. The extended joint connectedness measures largely confirm the traditional spillover estimates while providing a more theoretically grounded aggregate index. These findings shed light on the evolving integration of Eastern European stock markets, offering insights for portfolio diversification and financial stability monitoring.