Comparing the performance of downside arbitrage pricing theory (D-APT) and reward beta approach (RBA) in predicting stock returns in Tehran Stock Exchange

Activists in finance use various instruments to make optimal decision. One of the most important instruments is   stocks expected return that is predicted by some models. In this study two methods named reward beta approach and downside arbitrage pricing theory are investigated and compared. To test reward beta model sample period is divided into two sub-periods. First beta coefficients are estimated in the first period and then are used as explanatory variables to test the model in the second period. In investigating downside arbitrage pricing theory, only downside deviations of mean in calculating macroeconomic variables and stock returns are considered. To test this model, in order to predict beta coefficients and risk premiums of variables simultaneously, iterated nonlinear seemingly unrelated regression are used. The results indicate that during the period 1380-1397 in the case of stocks with lower market value, downside arbitrage pricing theory has better performance in predicting returns in Tehran Stock Exchange compared with reward beta approach.