[R] how to calculate standard error for the predicted value from geeglm?

[AB29]

Dear Sue, 

I am also having problems with this. As far as I can gather the predict
function will work with geeglm to give you predicted values from the model
but it does not produce the standard errors automatically. 

I have posted a similar question and have had no answers. However, I know
that there must be a way to do this?! 

Did you manage to find the answer? Or can anybody else help? 

Thanks 

Anna  



Li, Sue wrote:
> 
> Hello R-helpers,
> 
>  
> 
> I would like to calculate the standard error for the predicted value
> from geeglm.  
> 
> As an example, I would like to calculate the GEE mean of treatments and
> their standard error.  I first specified the model as 
> 
>  mod <- geeglm(resp ~ trt,
> data=dat,id=id,family=Gaussian,corstr="ar1",weights=weight)
> 
>  
> 
> Then I predicted the GEE mean and se using the following code
> 
> pred.mean <-
> predict(mod,data.frame(trt=factor(unique(dat$trt),levels=levels(dat$trt)
> )),se=T)
> 
>  
> 
> I got the error message "Error in XRinv^2 %*% rep(res.var, p) :
> non-conformable arguments".  
> 
>  
> 
> Thanks for your help,
> 
>  
> 
> Sue
> 
>  
> 
>  
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 
> 

-- 
View this message in context: http://r.789695.n4.nabble.com/how-to-calculate-standard-error-for-the-predicted-value-from-geeglm-tp3064560p3317287.html
Sent from the R help mailing list archive at Nabble.com.