I am doing research work on “Idiosyncratic volatility and stock return”.

I have calculated Idiosyncratic volatility with the help of Fama french three factor model. IV is defined as the standard deviation of the error term. We have estimated monthly IV using window of one year daily data. data is from July 2005 to June 2019.

I have taken the sample of 501 companies of BSE S&P 500 index of India and divided all of them in ten groups on the basis of IV of the month of June. Below I have presented the average monthly return of Equal weighted and Value weighted portfolio formed on the basis of IV. I have also run the Fama French three factor model on the return of these portfolios. Results- For equal weighted portfolio, average return are increasing from low to high portfolio but insignificant. But for High minus low (H-L) portfolio average return is negative and insignificant. Interc ept are negative for all portfolios and only for H-L portfolio it is significant. All the three independent variables are significant. For value weighted portfolios average returns are positive and insignificant for low and become negative and insignificant as we move towards high portfolios. The average return for a H-L portfolio, is negative and insignificant. Intercepts for all portfolios are negative but some intercepts are significant and insignificant for H-L portfolio. Now I am confused in interpreting these results please help me. After this I have also return the Fama MacBeth two stage cross section regression of Excess return on IV, Beta, ln (Market capitalization) and Book to market ratio. Which is giving all coefficients significant. There is positive coefficient for Idiosyncratic risk. Which means there is positive relation between Idiosyncratic risk and returns but my problem is, it is not clear from portfolio analysis as I have explained above. Please help me in interpreting these results.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.