Fama and French (Multiple Regression)

igsteo — Wed, 14 Apr 2021 08:19:12 GMT

/* Multiple Regression */

data fflib.temp2;
set fflib.famafrenchvariables;                    
by date;
array y[36] AAREIT_SP_Equity -- SUN_SP_Equity;         
do varNum = 1 to dim(y);
   varName = vname(y[varNum]);  
   Reit_rp = y[varNum];           
   output;
end;
drop AAREIT_SP_Equity -- SUN_SP_Equity;
run;
proc sort data=fflib.temp2;  by varName;  run;
ods output ParameterEstimates = fflib.ffparam5yrs
	(keep= varName Variable Estimate) ;
proc autoreg data=fflib.temp2;
     by varName;
     model Reit_rp = Mkt_rp smb hml/ dwprob;
     test Mkt_rp = 1;
     output out=fflib.famafrenchoutput p=p r=r ucl=u lcl=l alphacli=.10;
run;

I am having problems doing a multiple regression for fama and french. The above are my codes and these codes only regress mkt_rp. How should i amend this to regress mkt_rp, smb and hml.

thank you

Re: Fama and French (Multiple Regression)

sbxkoenk — Tue, 13 Apr 2021 14:46:46 GMT

Hello,

Do you mean that you do not get parameter estimates for the explanatory variables smb and hml?

In the picture of your dataset, the two variables (smb and hml) are constant. That's maybe / probably why.

It would be better to include the LOG. That is more telling than just the code.

Cheers,

Koen

topic Fama and French (Multiple Regression) in SAS Programming

Fama and French (Multiple Regression)

Re: Fama and French (Multiple Regression)