if I read that someone is using portfolio sorts to determine whether a factor is priced in the cross section ( risk premium ) is it the two-pass Fama-MacBeth regression? Is there a material that would clearly describe the steps perhaps?

It seems to me like everyone is having a cookbook, but I am not sure how to make the pie.

Thanks for any suggestions!

  • $\begingroup$ Hi: I would take a look at the link at the bottom of this comment. It probably doesn't answer your question exactly but the reference authored by Jegadeesh and Titmann, 1993, I think was one of the first papers to show evidence of factor alpha through the use of sorting. So, you should probably look at that one also. I don't know the field much so obviously there could be more modern approaches that I'm unaware of such as the one you mentioned. papers.ssrn.com/sol3/papers.cfm?abstract_id=2822686 $\endgroup$
    – mark leeds
    Jan 14 at 19:40
  • $\begingroup$ Waterboard the data hard enough to deduce hidden factors; and it will confess (to anything) ;-) Classic modern factor analysis. Max-respect to @markleeds; but the mania for ESG in active right now (given fee pressure from passive) is through the roof. Any "factor" that might be worth anything has a PE multiple of 100x ;-) 1000x if it cannot be measured ;-) $\endgroup$
    – demully
    Jan 15 at 2:10
  • $\begingroup$ Well stated by demully. I don't work with factors but, as he said, you play with anything long enough and you'll find something. One book that I don't have that gets good reviews and is on Amazon is called "Machine Learning For Factor Investing". If you work with R, you might want to check that out. I have no idea if it talks about sorting since there are obviously many ways to construct factors. It's possible that ML has sort of driven sorting away and it's less popular now ? $\endgroup$
    – mark leeds
    Jan 16 at 14:37


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.