topic Re: Logistic Regression Equation in SAS Procedures

Logistic Regression Equation

deleted_user — Thu, 27 Mar 2008 00:13:22 GMT

Dumb question, but what to ensure I have the application correct. When I have run my logistic regression and am staisfied with the results, how do I turn this into an equation to determine probabilities on new data? Is it simply taking the coefficient times the variable(s)?

Re: Logistic Regression Equation

deleted_user — Mon, 23 Jun 2008 19:16:17 GMT

proc logistic data=Neuralgia outmodel=sasuser.Model;
class Treatment Sex;
model Pain(event='Yes')= Treatment|Sex Age
/ selection=forward sle=.01;
run;

Suppose Treatment and Age are the effects selected for the final model. You can score a data set which does not contain the variable Sex since the effect Sex is not in the model that the scoring is based on.

proc logistic inmodel=sasuser.Model;
score data=Neuralgia(drop=Sex);
run;

Re: Logistic Regression Equation

Olivier — Tue, 24 Jun 2008 06:59:53 GMT

The Score statement in only valid in SAS 9.
For previous release, you can compute probabilites as :
1 / (1 + EXP(-1 * (coefficients * variables )))

(If you used a LOGIT link, of course.)

Regards
Olivier