I have a linear mixed-effects model from using the function. I'm trying now to check the assumptions of LME models, particularly absence of (multi)collinearity. I'm not sure of the best way to do this. It seems that testing the Variance Inflation Factor (VIF) is a good way to check this, although I'm not sure how to do that with the LME model. Any suggestions on how to do this? I've seen some functions like that seem to take a design matrix as the input, but not sure how to get that from the model. On a separate topic, could anyone provide any info on what the "lme.CoefficientCovariance" is? The documentation just says it's the "covariance of the estimated fixed-effects coefficients of the linear mixed-effects model," but I can't seem to find anything on what this actually means. Thanks!

Variance Inflation Factor (VIF) for Linear Mixed-Effects Model

Kyle

2018 5 月 24

0 回答

11 ビュー (30 日間)

2 投票

I have a linear mixed-effects model from using the fitlme function. I'm trying now to check the assumptions of LME models, particularly absence of (multi)collinearity. I'm not sure of the best way to do this. It seems that testing the Variance Inflation Factor (VIF) is a good way to check this, although I'm not sure how to do that with the LME model. Any suggestions on how to do this? I've seen some functions like this one that seem to take a design matrix as the input, but not sure how to get that from the model.

On a separate topic, could anyone provide any info on what the "lme.CoefficientCovariance" is? The documentation just says it's the "covariance of the estimated fixed-effects coefficients of the linear mixed-effects model," but I can't seem to find anything on what this actually means.

Thanks!