Estimating the Mean-Reverting Component in Stock Prices: A Cross-Country Comparison.

This paper investigates the mean-reverting component in stock prices for sixteen countries using a Kalman filter maximum likelihood estimation procedure to measure the transitory, permanent, and seasonal components. Evidence is provided supporting the mean reversion hypothesis that stock prices are not pure random walks: a statistically significant mean-reverting component is found in each country's stock prices. Nevertheless, for twelve of the sixteen countries, the transitory component does not explain more than five percent of the variation in stock prices. Copyright 1997 by Scottish Economic Society.