London Calling: Nonlinear Mean Reversion across National Stock Markets

This paper revisits empirical evidence of mean reversion of relative stock prices in international stock markets. We implement a strand of univariate and panel unit root tests for linear and nonlinear models of 18 national stock indices during the period 1969 to 2012. Our major findings are as follows. First, we find little evidence of linear mean reversion irrespective of the choice of a reference country. Employing panel tests yields the same conclusion once the cross-section dependence is controlled. Second, we find strong evidence of nonlinear mean reversion when the UK serves as a reference country, calling attention to the stock index in the UK. Choosing the US as a reference yields very weak evidence of nonlinear stationarity. Third, via extensive Monte Carlo simulations, we demonstrate a potential pitfall in using panel unit root tests with cross-section dependence when a stationary common factor dominates nonstationary idiosyncratic components in small samples.