Pastor–Stambaugh vs FFM

_ “Pastor–Stambaugh model adds a fourth factor to the FFM—compensation for the degree of liquidity of an equity investment.” _

To make sure that I got the two models right; the FFM includes three factors which are market, size and value

The Pastor-Stambaugh has four factors which are the Equity, size, industry risk and company specific risk.

Does this meant that the company specific risk is considered as the liquidity factor?

Pastor–Stambaugh model is the same as Fama French but adds Liquidity Factor

I realize this was a month ago but to answer your question - no it does not. FF and P-S both still include an error term at the end of the regression that represents idiosyncratic risk. The liquidity term is independent of that.

No this is not true, According to CFA text, Pastor and Stambaugh extended the FFM to encompass a compensation for the degree of liquidity of an equity investment and according to the two formulas in the curriculum there is no error term in FFM or Pastor formulas (Eq 11 / 12).

There most certainly is an error term - both models are constructed using ordinary least squares. Here are the original papers for each of the models:

Fama French (1992) http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.139.5892&rep=rep1&type=pdf

Pastor Stambaugh (2003) http://people.stern.nyu.edu/lpederse/courses/LAP/papers/TransactionCosts/PastorStam.pdf

We need to stick to CFAI material to answer the exam questions, specially that they didn’t mention error terms in these equations in Equity section (and I didn’t see it in any mock I have done) specially that it is 2 days now before the exam, Good luck !

Let me add something.

when we are using that equation for predictions, we ignore the error term since the regression model assumption expects the mean value of error term to be zero.

Agreed, you can only be tested by what’s in the material text but OP’s question was specifically whether the liquidity risk was equal to the company-specific risk. Conceptually understanding error terms will help you a ton in quant