[R] effect function (effects package)

[John Fox]

Dear Eileen,

You apparently misunderstand what the effect() function does; it computes
fitted values and their standard errors under the model (by default in a GLM
on the scale of the response) at specific combinations of values of the
predictors (e.g., letting two interacting predictors range over their
combinations of values while others are held to their means). Details are
provided in the references given in ?effect; see in particular the 2003
Journal of Statistical Software article at
<http://www.jstatsoft.org/v08/i15>. 

I hope this helps,
 John

--------------------------------
John Fox
Senator William McMaster
  Professor of Social Statistics
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
http://socserv.mcmaster.ca/jfox



> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Hlavka, Eileen
> Sent: February-04-12 1:56 AM
> To: r-help at r-project.org
> Subject: [R] effect function (effects package)
> 
> Dear all,
> 
> How does the effect() function in the effects package calculate
> effects and standard errors for glm quasipoisson models?  I was using
> effect() to calculate the impact of increasing x to e + epsilon, and
> then finding the expected percent change.  I thought that this effect
> (as a percentage) should be exp(beta*epsilon), where beta is the
> appropriate coefficient from the model, but that's not what I'm
> getting using the effect() output.
> 
> Sorry for the lack of example-it would require toy data etc. and seems
> unnecessary since my question is more conceptual.
> 
> Thanks,
> 
> Eileen
> 
> 
> 
> ______________________________________________________________________
> ____
> 
> This email message is for the sole use of the intended
> r...{{dropped:9}}
> 
> ______________________________________________
> 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.