This paper aims to introduce a new predictor of returns based on the long-run equilibrium relationship between dividends, prices and earnings (dpe for short). We compare results to the classical dividend-price (dp) and its modified version (mdp based on the cointegration relationship between dividends and prices). An investor who employs dpe and mdp improves in-sample forecasts by 49% and 43% respectively at the ten-year horizon, against dp which interprets merely 22% of time-varying expected returns. Additionally, out-of-sample (oos) performance testing shows that dp fails to generalise well, while mdp proves the strongest oos performer. Our proposed dpe contributes to empirical literature by resolving certain econometric issues and enhancing predictability findings in return forecasting. Also, this study introduces a simple modification in treating dividends, prices and earnings that can be easily replicated by practitioners in the field and can aid the work of financial analysts, investors, fellow researchers and portfolio managers.
No takes yet. Share an insight, caveat, or question.
Catherine Georgiou (2023) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: