This study aimed to present a model for portfolio risk premium assessment of companies listed in Tehran Stock Exchange. In order to achieve this purpose, monthly data of 150 companies listed in Tehran Stock Exchange during 2007-2017 was used. In this study, the predictive powers of FamaFrench three-factor model , Carhart four-factor model , Fama - French five-factor model , Brousseau five-factor model  and Roy and Shijin six-factor model  have been evaluated using variables used in the mentioned models and then an optimal model has been developed for portfolio risk assessment using stepwise regression. Findings showed that the Carhart four-factor model has higher predictive ability (48.3%) than other mentioned models in the Tehran Stock Exchange. According to the results of stepwise regression, seven variables have been selected as effective variables on portfolio risk premium. The explanatory power and predictive ability of the model developed in the Tehran Stock Exchange was 55.7% indicating higher predictive ability respect to previous models on portfolio risk premium. Investigation of the coefficients of the developed model showed that market risk premium, size factor, value factor, momentum factor and accounting quality factor have positive and significant effects on portfolio risk premium while investment factor and liquidity risk factor have significant negative impacts on portfolio risk premium.