[R] Longitudinal Weights in PLM package

[Millo Giovanni]

Dear Arne,

the inclusion of weights in the (mainly-GLS-related-) procedures in
'plm' is not obvious: at least, it is not to me. Maybe you might apply
the weights to the data before using them in estimation, which I have
done in the past, although it was then meant to reflect stratification,
not attrition: I admit my utter ignorance on this last particular
subject.

Unless you can resort to some pre-treatment of the data, you might be
better served with nlme/lme4. Should you need a quick "translation" of
syntax from 'plm' towards 'nlme', you van find it here, section 7:
http://www.jstatsoft.org/v27/i02

Best wishes,
Giovanni

---------- Original message --------------------

Message: 145
Date: Thu, 10 Feb 2011 01:51:33 -0800 (PST)
From: Arne <arne.warnke at googlemail.com>
To: r-help at r-project.org
Subject: [R] Longitudinal Weights in PLM package
Message-ID: <1297331493488-3298823.post at n4.nabble.com>
Content-Type: text/plain; charset=UTF-8


Hi all,

I a semi-beginner with R and I am working with the plm package to
examine a
longitudinal dataset. Each individual in this dataset has a longitudinal
weight for the probability that he or she remains in the sample. 

Unfortunately, I have not found an argument to use weights in the plm
function? I tried ?weights=? like in standard lm or in nlme or lm4 but
it
does not work. I asked the maintainer but I have not received a reply
yet. 

Therefore, I would like to ask you whether I can use plm with
longitudinal
weights or if I have to switch to other packages dealing with
longitudinal
data? 

Thank you in advance!

Best regards,

Arne  

----------- End original message ---------------------

Giovanni Millo
Research Dept.,
Assicurazioni Generali SpA
Via Machiavelli 4, 
34132 Trieste (Italy)
tel. +39 040 671184 
fax  +39 040 671160 

 
Ai sensi del D.Lgs. 196/2003 si precisa che le informazi...{{dropped:13}}