I am trying to investigate some trading strategies based on the Fama French 3-factor model, for which I assumed I need to use adjusted prices to account for dividends and splits. However, my regression give crazy outcomes if I use adjusted prices, and only 'normal' outcomes when I use the unadjusted regular closing prices. Is this normal, since this doesn't reflect 'true' returns? I am totally confused by this.

Thanks in advance!

  • $\begingroup$ If the two are wildly different, most likely you are computing returns incorrectly. $\endgroup$ – noob2 Mar 9 '16 at 14:48
  • $\begingroup$ I agree with @noob2. Calculate the correlation of the returns using prices vs. the returns using "adjusted" prices. It should be close to 99%. The standard deviations should be similar. The CAGR for the "adjusted" returns should be bigger than the CAGR of the price returns by roughly the dividend yield. $\endgroup$ – John Mar 9 '16 at 15:17
  • 1
    $\begingroup$ Check your adjusted data. Further, regressions are heavily effected by outliers or in your case possibly wrong adjustments. As in FF2008 I would winsor your data at the first and last percentiles. $\endgroup$ – Tim Mar 9 '16 at 19:15

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.