REGIME-SWITCHING STOCK RETURNS AND MEAN REVERSION

We estimate a well-specified two-state regime-switching model for Danish stock returns. The <p> model identifies two regimes which have low return-low volatility and high return-high <p> volatility, respectively. The low return-low volatility regime dominated, except in a few, short <p> episodes, until the beginning of the 70s whereas the 80s and 90s have been characterized by <p> high return and high volatility. We propose an alternative test of mean reversion which allows <p> for multiple regimes with potentially different constant and autoregressive terms and different <p> volatility. Using this test procedure we find mean reversion at 10% but not at 5% significance <p> level which is weaker evidence than produced by estimating a standard autoregressive model <p> for returns. Furthermore, when analyzing contributions of the two regimes we find that the <p> indication of mean reversion is due to the recent high return-high volatility regime only.