Abstract
After failing to reject the null of a unit root in the classical dividend-price ratio (dp), we show that a cointegrating relationship between dividends and prices exists. We then define the modified dividend-price ratio (mdp), as the long run trend deviation between log dividends and prices. Using S&P 500 data for the period 1926 to 2009, we show that mdp provides substantially improved forecasting results over the classical dp ratio. Actually, when both ratios are present in a multivariate setting, the classical dividend price ratio fails to bring any extra information about future returns. Thus, in the presence of the modified ratio the only thing left for the classical dividend-price ratio to predict is its forward looking value. Out of sample, while the classical dp ratio cannot outperform the “mean” benchmark for any useful horizon, an investor who employs the modified dp ratio will do better in forecasting 5and 7year returns with an between 52-54%. Further, with 1-year and 3-year of 12% and 39% respectively, mdp addresses a major weakness in dp, namely its presumed inability in revealing business cycle variation in expected returns and equity risk-premia. We show that the gain of our modified ratio in forecasting returns is mainly due to its enhanced ability to forecast their risk free component, and argue that it can be considered as a de-noising of the classical ratio. * Department of Economics, School of Law and Economics, Aristotle University of Thessaloniki. ineokosm@econ.auth.gr † Division of Business Administration, School of Law and Economics, Aristotle University of Thessaloniki. polimenis@econ.auth.gr ‡ Corresponding author.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have