R help - Dec 2005 - glmmPQL and variance structure

Dear listers,

glmmPQL (package MASS) is given to work by repeated call to lme. In the 
classical outputs glmmPQL  the Variance Structure is given  as " fixed 
weights,  Formula: ~invwt".  The script shows that the function 
varFixed() is used, though the place where 'invwt' is defined remains 
unclear to me.  I wonder if there is an easy way to specify another 
variance structure (eg varPower, etc..), preferably using an lme object 
of the varFunc classes ? Some trials show that the 'weights' argument of
glmmPQL is just the same as in glm (which is clearly stated in the help) 
and I wonder actually, if not a nonsense, how to pass eg a 'weights' 
arguments as used in lme (eg weights=varPower()) to specify a variance 
function (in the same way as a correlation structure can be passed easy).

Thanks in advance for any hint,

Patrick

Have you received a reply to this post?  I haven't seen one.  I don't 
have an answer for you, but if you'd still like help from this list, I 
suggest you prepare the simplest possible toy example that you can 
conceive and send it to this list, restating your question in terms of 
that example.  You question indicates you've read the code for glmmPQL 
and seem to know enough to experiment with modifying the "glmmPQL"
code
or with extracting crucial pieces to make part of the simplified example 
illustrating your question (consistent with the posting guide, 
"www.R-project.org/posting-guide.html").

	  If you are not sure what "glmmPQL" does, you can say
"debug(glmmPQL)"
before executing a command that invokes "glmmPQL".  That will open a 
browser that will allow you to look at and change anything in the 
environment of "glmmPQL" before and after any command;  if a command 
commits a fatal error, you will be evicted from "glmmPQL" and will
have
to start over.  This is the quickest way I know to understand and debug 
R code.

	  hope this helps.
	  spencer graves

Patrick Giraudoux wrote:
> Dear listers,
> 
> glmmPQL (package MASS) is given to work by repeated call to lme. In the 
> classical outputs glmmPQL  the Variance Structure is given  as " fixed
> weights,  Formula: ~invwt".  The script shows that the function 
> varFixed() is used, though the place where 'invwt' is defined
remains
> unclear to me.  I wonder if there is an easy way to specify another 
> variance structure (eg varPower, etc..), preferably using an lme object 
> of the varFunc classes ? Some trials show that the 'weights'
argument of
> glmmPQL is just the same as in glm (which is clearly stated in the help) 
> and I wonder actually, if not a nonsense, how to pass eg a
'weights'
> arguments as used in lme (eg weights=varPower()) to specify a variance 
> function (in the same way as a correlation structure can be passed easy).
> 
> Thanks in advance for any hint,
> 
> Patrick
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
-- 
Spencer Graves, PhD
Senior Development Engineer
PDF Solutions, Inc.
333 West San Carlos Street Suite 700
San Jose, CA 95110, USA

spencer.graves at pdf.com
www.pdf.com <http://www.pdf.com>
Tel:  408-938-4420
Fax: 408-280-7915

Dear listers,

On the line of a last (unanswered) question about glmmPQL() of the 
library MASS, I am still wondering if it is possible to pass a variance 
structure object  to the call to lme() within the functions (e.g. 
weights=varPower(1), etc...). The current weights argument of glmmPQL is 
actually used for a call to glm -and not for lme). I have tried to go 
through the code, and gathered that the variance structure passed to the 
call to lme()  was:

mcall$weights <- quote(varFixed(~invwt))

and this cannot be modified by and argument of glmmPQL().

I have tried to modify the script a bit wildly  and changed varFixed 
into VarPower(~1), in a glmmPQL2 function. I get the following error:

 > glmmPQL2(y ~ trt + I(week > 2), random = ~ 1 | ID,
+ family = binomial, data = bacteria)
iteration 1
Error in unlist(x, recursive, use.names) :
        argument not a list

I get the same error whatever the change in variance structure on this line.

Beyond this I wonder why variance structure cannot be passed to lme via 
glmmPQL...

Any idea?

Patrick Giraudoux