2

I am an undergraduate student . In this level , we aren't taught Multilevel Logistic Regression. But my project topic is Multilevel Logistic Regression and I am working with only two-level . What I know only to use the glmer() function in R to estimate the fixed and random effect parameters . But I don't know , when to use which types of residuals , though R has given the reference Davison, A. C. and Snell, E. J. (1991) Residuals and diagnostics for the usages of each type , I didn't get the pdf of the reference .

Also I haven't any theoretical background of estimation procedure .

Would you please give me some online reference so that I can conduct my project issue sample size determination in Multilevel Logistic Regression fairly well ?

Thanks ! Regards .

user81411
  • 731
  • 1
  • 7
  • 14
  • 1
    Why do you want to use residuals from a logistic regression analysis, more or less a multilevel model? You hardly can do any of the diagnostic checking in either case. – AdamO Aug 18 '15 at 22:35

1 Answers1

1

I think you can refer to residuals.merMod(). The default one for glmer(), deviance, makes more sense. More examples are shown in this related question: What do the residuals in a logistic regression mean?

The default residual type varies between lmerMod and glmerMod objects: they try to mimic residuals.lm and residuals.glm respectively. In particular, the default type is "response", i.e. (observed-fitted) for lmerMod objects vs. "deviance" for glmerMod objects. type="partial" is not yet implemented for either type.

Community
  • 1
Randel
  • 6,199
  • 4
  • 39
  • 65