I am writing a mid-term thesis on the Fama-French factor model.

I have built 5 portfolios sorted by the Book-to-Market ratio. The first portfolio is the lowest-BM group and the last portfolio is the highest-BM group. The interception and the coefficients are all insignificant for the first two portfolios (low-BM). The beta_HML is significant for the last three portfolios(high-BM). I also set a group which is the difference between the last and the first portfolio, the coefficients are all insignificant.

How can I interpret such a result?

Does the model fail to explain excess returns of the first two groups?

Does the result mean that the 3-factor model couldn't explain the returns difference between the last group and the first group?

Thank you!

  • $\begingroup$ It would we helpful if you could clarify your stock universe (which market and what stocks are excluded) and some summary statistics on the portfolio returns. $\endgroup$ – skoestlmeier Aug 23 '19 at 21:05
  • $\begingroup$ @skoestlmeier moin! The stocks are in the U.S. market and no specific stocks are excluded. I just trimmed 1% top and 1% bottom of the BM ratio. The average returns across the time for the first portfolio are around 0.12, the last 0.24. The returns do increase along with the increase in BM ratio. According to the rationale of FF, all my portfolios should have significant beta_HML, right? $\endgroup$ – MiaZimmermann Aug 24 '19 at 9:24
  • $\begingroup$ Short remarks in lack of time for a full answer: Typically, financial stocks (SIC bet. 6,000 - 6,999) are excluded, as well as observations with negative book-value. The US value-effect has it roots mainly within the portfolio of the highest book-to-market ratios. Not all portfolios have to have excess returns which are significant different from zero, but the high minus low portfolio commonly has a significant (pos.) excess return. Besides Fama/French, you may take a look on the chapter "value effect" in Bali/Engle/Murray (2016), Empirical Asset Pricing. $\endgroup$ – skoestlmeier Aug 26 '19 at 7:38

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.