R help - Mar 2000 - Corstr in the Gee (Generalized Estimation Equation) arguments?

Dear all:
Y=a+bX1+cX2
In the Gee (Generalized Estimation Equation) arguments:
The arument Corstr has sveral choices:
            "independence" "fixed" "stat_M_dep"
"non_stat_M_dep"
             "exchangeable" "AR-M" "unstructured" 
What does each term mean?
How do I choose among them?
How do I know the correlation structure of my repeated measures data?
Can non-normally distributed X1, X2 be used?
Thank you very much.

Sincerely,
Jinn-Yuh Guh
Kaohsiung Medical Colege
jyuh at mail.nsysu.edu.tw
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
https://stat.ethz.ch/pipermail/r-help/attachments/20000318/4617966b/attachment.html

On Sat, 18 Mar 2000, Jinn-Yuh Guh wrote:
> Dear all:
> Y=a+bX1+cX2
> In the Gee (Generalized Estimation Equation) arguments:
> The arument Corstr has sveral choices:
>             "independence" "fixed"
"stat_M_dep" "non_stat_M_dep"
>              "exchangeable" "AR-M"
"unstructured"
> What does each term mean?
> How do I choose among them?
> How do I know the correlation structure of my repeated measures data?
> Can non-normally distributed X1, X2 be used?
> Thank you very much.
The distribution of X1,X2 is irrelevant, as for most regression
methods. The distribution of Y can be non-Normal, the family= argument
works the same way as for glm.

There isn't a magic solution to knowing the correlation of your data. You
should read 

	@article{gee-ZL,author={Scott L. Zeger and Kung-Yee Liang},
        	  title={Longitudinal Data Analysis for Discrete and
                  Continuous Outcomes}, journal={Biometrics},
                  year={1986}, volume=42, pages={121--130}}
                       
and/or

	@Book{diggle,
	author = "Peter J. Diggle and Kung-Yee Liang and Scott L. Zeger",
  	title = "Analysis of Longitudinal Data",
  	publisher = "Oxford University Press",
  	year = 1994,
  	series = "Oxford Statistical Science Series",
  	address = "Oxford" }
   
or other references on the subject.


	-thomas

Thomas Lumley
Assistant Professor, Biostatistics
University of Washington, Seattle

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._