topic Re: how to calculate VIF using PROC GLM? in Statistical Procedures

how to calculate VIF using PROC GLM?

bhr-q — Tue, 10 Sep 2024 12:20:19 GMT

Hello all,

Is there any way to check multicollinearity using proc GLM?

proc glm data=tmp  ; 
class study CNS stk_loc sex;
model fim =study sex age CNS stk_loc /solution ss3 ;
run;

The dependent variable is continuous

All the independent variables are categorical except age which is continuous

I am familiar with VIF in proc reg, but needed to create dummy variables.

any input appreciated

Re: how to calculate VIF using PROC GLM?

PaigeMiller — Tue, 10 Sep 2024 12:46:28 GMT

Use PROC GLMMOD to obtain the x matrix used by PROC GLM. Then run PROC REG with the VIF option, using the output of PROC GLMMOD as input to PROC REG.

Or maybe this: https://stackoverflow.com/questions/77531415/sas-basic-analysis-problems

Re: how to calculate VIF using PROC GLM?

sbxkoenk — Tue, 10 Sep 2024 13:14:41 GMT

Concurring with @PaigeMiller .

In PROC GLMMOD , you can use the OUTDESIGN= option.
In PROC GLMMOD , you can use ODS to create the design matrix data set.

The results are equivalent, but the columns of the data set produced by ODS have names that are directly related to the names of their corresponding effects.

Here's an example of the latter option:

SAS Help Center: Example 49.2 Factorial Screening

ods output DesignPoints = DesignMatrix;
proc glmmod data=Screening;
   model y = a|b|c|d|e@2;
run;

proc reg data=DesignMatrix;
   model y = a--d_e;
   model y = a--d_e / selection = forward
                      details   = summary
                      slentry   = 0.05;
run;
QUIT;

Koen

Re: how to calculate VIF using PROC GLM?

Ksharp — Wed, 11 Sep 2024 01:45:08 GMT

1) You could use PROC GLMSELECT to eliminate these multicollinearity variables.

2)You could use PROC GENMOD + CORRB option to check the correlation between these estiamte coefficient.

proc genmod data=sashelp.heart  ; 
class status bp_Status sex;
model weight =status bp_Status sex height / corrb ;
quit;

Re: how to calculate VIF using PROC GLM?

bhr-q — Wed, 11 Sep 2024 02:22:14 GMT

Thanks for your answer, it was interesting to get the tolerance using Proc GLM, but when I used Proc GLM with the tolerance option it didn't show me any tolerance.

Re: how to calculate VIF using PROC GLM?

bhr-q — Wed, 11 Sep 2024 02:24:21 GMT

Thanks for your answer, it was helpful.

Re: how to calculate VIF using PROC GLM?

sbxkoenk — Wed, 11 Sep 2024 08:27:42 GMT

@bhr-q wrote:
Thanks for your answer, it was interesting to get the tolerance using Proc GLM, but when I used Proc GLM with the tolerance option it didn't show me any tolerance.

Weird.

Here's some info on the tolerance output in PROC GLM :

The Type 1 tolerance of a parameter is the tolerance for this parameter with respect to the preceding parameters in the model.
The Type 2 tolerance of a parameter is the tolerance for this parameter with respect to all other parameters in the model.

The TYPE 2 Tolerance is consistent with the TOL option on the MODEL statement in PROC REG, which is 1/VIF.

It is your choice which tolerance to use.

Koen

Re: how to calculate VIF using PROC GLM?

PaigeMiller — Wed, 11 Sep 2024 11:05:16 GMT

@Ksharp wrote:

1) You could use PROC GLMSELECT to eliminate these multicollinearity variables.

2)You could use PROC GENMOD + CORRB option to check the correlation between these estiamte coefficient.
proc genmod data=sashelp.heart  ; 
class status bp_Status sex;
model weight =status bp_Status sex height / corrb ;
quit; 

The problem with CORRB is that it only looks for pairwise correlations. Maybe for some data sets, that's fine but it will miss more complicated types of correlations. VIF (and Tolerance) looks for correlation with the combination of all other parameters in the model.

Re: how to calculate VIF using PROC GLM?

Ksharp — Thu, 12 Sep 2024 00:30:13 GMT

Paige,
I know what you are talking about (linear combination of multiple variables).
But I don't think there is a problem by deleting/detecting a variable one by one(the linear combination of multiple variables must be high correlated with one of these variables ).
If you are not agree with that , you could try PROC GLMSELECT or HPGENSELECT that would take care of your consideration.