Regression logistic

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

abis86
Posts: 1
Joined: Tue Apr 26, 2016 10:25 am

Regression logistic

Postby abis86 » Tue Apr 26, 2016 10:28 am

Goodmorning everyone,

I am conducting a logistic regression to study the possible prediction of a tumor on the basis of a series of parameters recorded on patients two years before their onset ascertained.

The point is that doing binary regression with SPSS, I get that the variables that remain valid in the equation, change when I change between Forward / Backward model (wald etc).

How is it possible?

thank you so much!
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Regression logistic

Postby GerineL » Tue May 10, 2016 9:27 am

It is a different estimation method. With the forward method, you keep adding variables until no more improvement in the model is measured. With backward elimination, you start with all variables and eliminate until no more improvement is possible.

Who is online

Users browsing this forum: No registered users and 4 guests

cron