Estimation of US Stock Prices Based on Dividend Discount Models

This study is conducted to analyze the quality of estimates of US stocks from S&P 500 index by Dividend Discount Models over the past 11 years, including 2020, which was affected by the COVID-19 pandemic. Three specifications of DDM are used to obtain these estimates, including its primary ones and complex variant with more than five financial indicators. In order to check proximity of stock prices to predictions, statistical measurements such as RMSD, MAE and MAPE are used for the entire research sample. In addition, the analysis of models’ quality is carried out using regressions, where the discount rate is an unknown coefficient. The novelty of this study lies in the fact that a comparison of three DDM variations was made for the last decade, where a significant difference was found between the basic and complex models. The last model shows much better approximations to real data prices. Another important finding is that there is no significant impact of the COVID-19 pandemic on the quality of models estimates, potentially making it reliable to use not only during stable periods, but also in similar shocks caused by pandemic. There results may be useful for further DDM implementation both by scientists and investors.