Abstract:
The main goal of the present study is testing asymmetric risk pricing and comparing it
with pricing of traditional risk measures in Tehran Stock Market. Accordingly, a sample
consisting of 101 companies listed in Tehran Stock Market during 2002-2013 went under
investigation. In order to test asymmetric risk pricing, regression model of panel data was
applied. The results revealed a positive and significant relationship between traditional
measures (Standard Deviation and Semi Standard Deviation) and asymmetric risk
measures (parametric VaR, HR risk, historical VaR, and historical HR) and expected
return. Therefore, in addition to the significant correlation between risk and return,
pricing model based on asymmetric risk and traditional risk was approved, too. Again, it
was shown that controlling the effect of variables such as financial leverage, firm size,
book-to-market ratio of equity (B/M) and liquidity, momentum and inverse is not able to
change the direction of the relationship. Furthermore, the explanatory power of traditional
and asymmetric risk criteria are the same.
Machine summary:
"In previous studies, such as Fama and French (1992), and Haffman and Moll (2012), the effect of variables such as financial leverage, firm size, equity book value to market value (B/M), and liquidity on the relationship between volatility and expected return has been confirmed.
Accordingly, like Haffman and Moll (2012), the present research studies the effect of the mentioned variables on the relationship between asymmetric risk measures and expected return through Panel Data Regression model.
6 Findings In the present study, the relationship between asymmetric and symmetrical risk measures and expected return was investigated using regression model of panel data; meantime, the effect of variables like financial leverage, firm size, book value to market value, liquidity, momentum and reversal of the relationship between asymmetric risk and expected return were controlled.
Table 6 The Relationship between Risk Measures and Expected Return after Considering Control Variables Variable Model 7 Model 8 Model 9 Model 10 Model 11 Model (View the image of this page)*, **, and *** show level of significance at 90%, 95%, and 99%, the numbers in parentheses are t value, respectively.
Table 7 The Relationship between Risk Measures and Expected Return after Consideration of the Effect of Momentum and Reversal (View the image of this page) *, **, and *** respectively show level of significance at 90%, 95%, and 99%, the numbers in parentheses are t value.
The findings of this study revealed a significant positive relationship between traditional measures (SD, semi SD) and asymmetric risk measures (parametric VaR, parametric HR risk, historical VaR, historical HR), and expected return by using the regression model of panel data."