Timeline for Omitting more than one factor from a factor model
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 10 at 19:43 | comment | added | Richard Hardy | @KaiSqDist, I explain in the quoted posts why your suggestion does not make sense if we care about model validity (though not so much explanatory power). Or actually, I quote John Cochrane's explanation. He calls this a common misconception. This has to do with estimation of betas under different models. | |
| Dec 10 at 16:09 | comment | added | KaiSqDist | If SMB and HML are statistically significant, wouldn't it make sense to keep the factors in the model? Not quite understanding what is the purpose of omitting these 2 factors. Because if you omitted SMB and HML $despite$ their statistical significance, the model (CAPM in this case) and the coefficient of MKT would be misspecified. | |
| Dec 10 at 16:00 | comment | added | Richard Hardy | @phdstudent, I do not think this is their primary purpose, and it is not obvious to me they would answer the question, but maybe? What would be the logic? | |
| Dec 10 at 15:52 | comment | added | phdstudent | Is that why one runs Fama MacBeth regressions? | |
| Dec 10 at 15:35 | history | asked | Richard Hardy | CC BY-SA 4.0 |