[R-sig-ME] Unstandardizing GLMM Regression Coefficients

[Ben Bolker]

On 16-01-18 01:52 AM, Timothy Lau wrote:
> Hello,
>
> lme4 mentioned that I should consider rescaling some of my predictor
> variables because of the scale differences:
>
> *fit warnings:*
> *Some predictor variables are on very different scales: consider rescaling*
>
>
>
> I have since been trying to expand Ben Bolker's function:

A brief note -- I've updated the end of the StackOverflow answer you 
reference 
<http://stackoverflow.com/questions/23642111/how-to-unscale-the-coefficients-from-an-lmer-model-fitted-with-a-scaled-respon> 
to read:

 > If you scale only the predictors and not the response, then submit 
c(0,mean(predictors)) for m and c(1,sd(predictors)) for s

   that is, you shouldn't need to modify the given function very much ...

   Ben Bolker