Stock Returns Under Different Market Regimes: An Application of Markov Switching Models to 24 European Indices

This paper studies the different modes of operation of European stock markets. Using data on 24 European indices over a period of 15 years, we show that these can be well represented by a Hidden Markov Model with two regimes that roughly correspond to bull and bear markets. We further estimate regime parameters and show that the alternate regimes have very different risk-return tradeoffs with clear implications for portfolio management. Corresponding transition probability matrices show the remarkable persistence of states and give a possible quantitative estimate of the degree of inertia in financial markets. Regime-switching coordination across markets is further examined, showing that moments of correlations are followed by idiosyncratic episodes and thus, risk diversification through regime arbitrage is possible.