F-test to compare two models

Reply
Super Contributor
Super Contributor
Posts: 440

F-test to compare two models

Hi guys,

I have two models that i need to compare by  conducting  a F-test but got no clue how

y=b0+b1x1+b2x2

y=b0+b1x1+b2x2+b3x3

I did google and came across this :

proc reg data = mydata ;

        model y = x1 x2 x3 ;

        test x3 = 0;

run ;

Is this something that i need to do,anyone please?

Respected Advisor
Posts: 4,646

Re: F-test to compare two models

I would suggest to take some basic statistical training to learn why the test statement is redundant and how to interpret the analysis of variance of your model to determine if variable x3 is useful to predict y.

PG

PG
Super Contributor
Super Contributor
Posts: 440

Re: F-test to compare two models

my second model is  actually more complicated than it  shows here in my post. For example lets say it is b0+b1x1+b2X2+....+b6x6

and i know how to test the null hypothesis based on the global F test regardless of whether the individual variables (x3-x6) contribute to the model or not but are  satisfactory F and p  results  for the second model enough to chose it over the  1st model?
so there is no way i can compare these models and based on F and p determine if the complete model is good to go or not without running them separately ?

Respected Advisor
Posts: 2,655

Re: F-test to compare two models

A likelihood ratio test seems in order here.  Fit the full and reduced models in PROC MIXED, and get the difference in -2 log likelihood values.  This will be distributed as a chi square with degrees of freedom equal to the number of parameters "deleted" by reducing the model.

Steve Denham

Ask a Question
Discussion stats
  • 3 replies
  • 634 views
  • 8 likes
  • 3 in conversation