Investigating Predictability of Different "Forms of Return" in Tehran Stock Exchange: Some Rolling Regressions-based Evidence

Abstract

This paper has provided "out of sample" evidence of stock returns predictability in Tehran Stock Exchange. 68 qualified companies over the period from 2002 to 2015 were selected and for five different "forms of returns", five superior predictive models have been designed by applying "General to specific" approach of modeling technique. Then "out of sample" analysis, based on rolling regressions, has been used to test the validation of the designed models. The result showed that all designed models have sufficient "out of sample" validity and the aggregate returns have a higher predictability level.