1

I have generated a glm model with 20 or so predictors. I have carried out stepwise regression(forward and backwards selection) to identify the important predictor variables. My final model has 7 predictor variables. The results are below:

Coefficients:
                       Estimate Std. Error z value Pr(>|z|)    
(Intercept)           -7.888112   0.859847  -9.174  < 2e-16 ***
age                    0.028529   0.009212   3.097  0.00196 ** 
bmi                    0.095759   0.015265   6.273 3.53e-10 ***
surgery11              0.923723   0.524588   1.761  0.07826 .  
surgery21              1.607389   0.600113   2.678  0.00740 ** 
surgery31              1.544822   0.573972   2.691  0.00711 ** 
cvd1                   0.624692   0.290005   2.154  0.03123 *  
rt_1                   -0.816374   0.353953  -2.306  0.02109 *

I want to see if multi-colinearity exists, so I have used the VIF function from the car package. My understanding is that VIF is used for linear models, so I was wondering whether it can be used in glm (logistic) models?

However, I am unsure whether VIF can be applied to a logistic model? I have the results below for VIF:

vif(model_logistic)

                 age                  bmi             surgery1              surgery2 
            1.046694             1.008971             6.256793             3.658226 
             surgery_3                  cvd            rt_1 
            4.660840             1.038339             1.144582
kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
HKJ3
  • 11
  • 1
  • Does this answer your question? [In plain language, why is there no VIF for binary outcome regression models?](https://stats.stackexchange.com/questions/112971/in-plain-language-why-is-there-no-vif-for-binary-outcome-regression-models) Note that stepwise regression is [problematic](https://www.stata.com/support/faqs/statistics/stepwise-regression-problems/) and [multicollinearity need not be an issue](https://stats.stackexchange.com/questions/555145/ridge-regression-for-multicollinearity-and-outliers/555163#555163). – Dave Dec 08 '21 at 22:30

0 Answers0