Re: [R] Variance Components in R

From: Iuri Gavronski <iuri_at_proxima.adm.br>
Date: Sun 20 Aug 2006 - 22:13:38 EST

Reading Bates' article on R News, I see that random effects require a grouping variable. As, by convention, all variables in G-studies are supposed random, what could be a grouping variable in that case? I see that the model I wrote before (if ever ran...) would take all effects as fixed.

Is it possible to use lmer() without fixed effects?

Anything would help.

Iuri.

On 8/18/06, Iuri Gavronski <iuri@ufrgs.br> wrote:
>
> Harold,
> I don't have a grouping variable. And yes, persons can be an important
> source of variance, and they are the "resp" variable. "rating" is the
> response.variable in the model you specified below. "aov" perhaps could
> give me distorted results, because of unbalanced data (what estimation
> method it uses? ANOVA?): not all respondents evaluated all stores.
> I have five variables: resp (persons, the respondents), aspect (the
> "construct"), item (the "question"), chain (the "store" the person is
> rating) and sector (the economic sector where chain belongs, e.g.
> groceries). And one response, "rating".
>
> The model would be?
>
> fm <- lmer(rating ~ resp + aspect + item + chain + sector + sector*resp +
> sector*aspect + sector*item + chain*resp + chain*aspect + chain*item +
> resp*aspect + resp*item + sector*resp*aspect + sector*resp*item +
> chain*resp*aspect)
>
>
> Regards,
>
> Iuri.
>
> On 8/17/06, Doran, Harold <HDoran@air.org> wrote:
> >
> > Iuri:
> >
> > Here is an example of how a model would be specified using lmer using a
> > couple of your factors:
> >
> > fm <- lmer(response.variable ~ chain*sector*resp
> > +(chain*sector*resp|GroupingID), data)
> >
> > This will give you a main effect for each factor and all possible
> > interactions. However, do you have a grouping variable? I wonder if aov
> > might be the better tool for your G-study?
> >
> > As a side note, I don't see that you have a factor for persons. Isn't
> > this also a variance component of interest for your study?
> >
> > ------------------------------
> > *From:* prof.iuri@gmail.com [mailto:prof.iuri@gmail.com] *On Behalf Of *Iuri
> > Gavronski
> > *Sent:* Thursday, August 17, 2006 1:26 PM
> > *To:* Doran, Harold
> >
> > *Cc:* r-help@stat.math.ethz.ch
> > *Subject:* Re: [R] Variance Components in R
> >
> > I am trying to replicate Finn and Kayandé (1997) study on G-theory
> > application on Marketing. The idea is to have people evaluate some aspects
> > of service quality for chains on different economy sectors. Then, conduct a
> > G-study to identify the generalizability coefficient estimates for different
> > D-study designs.
> > I have persons rating 3 different items on 3 different aspects of
> > service quality on 3 chains on 3 sectors. It is normally assumed on
> > G-studies that the factors are random. So I have to specify a model to
> > estimate the variance components of CHAIN SECTOR RESP ASPECT ITEM, and
> > the interaction of SECTOR*RESP SECTOR*ASPECT SECTOR*ITEM CHAIN*RESP
> > CHAIN*ASPECT CHAIN*ITEM RESP*ASPECT RESP*ITEM SECTOR*RESP*ASPECT
> > SECTOR*RESP*ITEM CHAIN*RESP*ASPECT. '*' in VARCOMP means a crossed design.
> > Evaluating only the two dimensions interactions (x*y) ran in few minutes
> > with the full database. Including three interactions (x*y*z) didn't complete
> > the execution at all. I have the data and script sent to a professor of the
> > department of Statistics on my university and he could not run it on either
> > SPSS or SAS (we don't have SAS licenses here at the business school, only
> > SPSS). Nobody here at the business school has any experience with R, so I
> > don't have anyone to ask for help.
> > Ì am not sure if I have answered you question, but feel free to ask it
> > again, and I will try to restate the problem.
> >
> > Best regards,
> >
> > Iuri
> >
> > On 8/17/06, Doran, Harold <HDoran@air.org> wrote:
> >
> > > This will (should) be a piece of cake for lmer. But, I don't speak
> > > SPSS. Can you write your model out as a linear model and give a brief
> > > description of the data and your problem?
> > >
> > > In addition to what Spencer noted as help below, you should also check
> > > out the vignette in the mlmRev package. This will give you many examples.
> > >
> > > vignette('MlmSoftRev')
> > >
> > >
> > >
> > >
> > > ------------------------------
> > > *From:* prof.iuri@gmail.com [mailto:prof.iuri@gmail.com] *On Behalf Of
> > > *Iuri Gavronski
> > > *Sent:* Thursday, August 17, 2006 11:16 AM
> > > *To:* Doran, Harold
> > >
> > > *Subject:* Re: [R] Variance Components in R
> > >
> > > 9500 records. It didn`t run in SPSS or SAS on Windows machines, so I
> > am trying to convert the SPSS script to R to run in a RISC station at the
> > university.
> >
> > On 8/17/06, Doran, Harold <HDoran@air.org> wrote:
> >
> > > Iuri:
> >
> > The lmer function is optimal for large data with crossed random effects.
> > How large are your data?
> >
> > > -----Original Message-----
> > > From: r-help-bounces@stat.math.ethz.ch
> > > [mailto: r-help-bounces@stat.math.ethz.ch] On Behalf Of Iuri Gavronski
> >
> > > Sent: Thursday, August 17, 2006 11:08 AM
> > > To: Spencer Graves
> > > Cc: r-help@stat.math.ethz.ch
> > > Subject: Re: [R] Variance Components in R
> > >
> > > Thank you for your reply.
> > > VARCOMP is available at SPSS advanced models, I'm not sure
> > > for how long it exists... I only work with SPSS for the last
> > > 4 years...
> > > My model only has crossed random effects, what perhaps would
> > > drive me to lmer().
> > > However, as I have unbalanced data (why it is normally called
> > > 'unbalanced design'? the data was not intended to be
> > > unbalanced, only I could not get responses for all cells...),
> > > I'm afraid that REML would take too much CPU, memory and time
> > > to execute, and MINQUE would be faster and provide similar
> > > variance estimates (please, correct me if I'm wrong on that point).
> > > I only found MINQUE on the maanova package, but as my study
> > > is very far from genetics, I'm not sure I can use this package.
> > > Any comment would be appreciated.
> > > Iuri
> > >
> > > On 8/16/06, Spencer Graves <spencer.graves@pdf.com > wrote:
> > > >
> > > > I used SPSS over 25 years ago, but I don't recall
> > > ever fitting a
> > > > variance components model with it. Are all your random
> > > effects nested?
> > > > If they were, I would recommend you use 'lme' in the 'nlme' package.
> > > > However, if you have crossed random effects, I suggest you
> > > try 'lmer'
> > > > associated with the 'lme4' package.
> > > >
> > > > For 'lmer', documentation is available in Douglas
> > > Bates. Fitting
> > > > linear mixed models in R. /R News/, 5(1):27-30, May 2005
> > > > (www.r-project.org -> newsletter). I also recommend you try the
> >
> > > > vignette available with the 'mlmRev' package (see, e.g.,
> > > > http://finzi.psych.upenn.edu/R/Rhelp02a/archive/81375.html ).
> >
> > > >
> > > > Excellent documentation for both 'lme' (and indirectly for
> > > > 'lmer') is available in Pinheiro and Bates (2000)
> > > Mixed-Effects Models
> > > > in S and S-Plus (Springer). I have personally recommended
> > > this book
> > > > so many times on this listserve that I just now got 234 hits for
> > > > RSiteSearch("graves pinheiro"). Please don't hesitate to pass this
> > > > recommendation to your university library. This book is
> > > the primary
> > > > documentation for the 'nlme' package, which is part of the
> > > standard R
> > > > distribution. A subdirectory "~library\nlme\scripts" of your R
> > > > installation includes files named "ch01.R", "ch02.R", ...,
> > > "ch06.R",
> > > > "ch08.R", containing the R scripts described in the book. These R
> > > > script files make it much easier and more enjoyable to study that
> > > > book, because they make it much easier to try the commands
> > > described
> > > > in the book, one line at a time, testing modifications to check you
> > > > comprehension, etc. In addition to avoiding problems with
> > > > typographical errors, it also automatically overcomes a few
> > > minor but
> > > > substantive changes in the notation between S-Plus and R.
> > > >
> > > > Also, the "MINQUE" method has been obsolete for over
> > > 25 years.
> > > > I recommend you use method = "REML" except for when you want to
> > > > compare two nested models with different fixed effects; in
> > > that case,
> > > > you should use method = "ML", as explained in Pinheiro and
> > > Bates (2000).
> > > >
> > > > Hope this helps.
> > > > Spencer Graves
> > > >
> > > > Iuri Gavronski wrote:
> > > > > Hi,
> > > > >
> > > > > I'm trying to fit a model using variance components in R, but if
> > > > > very new on it, so I'm asking for your help.
> > > > >
> > > > > I have imported the SPSS database onto R, but I don't know how to
> > > > > convert the commands... the SPSS commands I'm trying to
> > > convert are:
> > > > > VARCOMP
> > > > > RATING BY CHAIN SECTOR RESP ASPECT ITEM
> > > > > /RANDOM = CHAIN SECTOR RESP ASPECT ITEM
> > > > > /METHOD = MINQUE (1)
> > > > > /DESIGN = CHAIN SECTOR RESP ASPECT ITEM
> > > > > SECTOR*RESP SECTOR*ASPECT SECTOR*ITEM CHAIN*RESP
> > > > > CHAIN*ASPECT CHAIN*ITEM RESP*ASPECT RESP*ITEM
> > > > > SECTOR*RESP*ASPECT SECTOR*RESP*ITEM
> > > CHAIN*RESP*ASPECT
> > > > > /INTERCEPT = INCLUDE.
> > > > >
> > > > > VARCOMP
> > > > > RATING BY CHAIN SECTOR RESP ASPECT ITEM
> > > > > /RANDOM = CHAIN SECTOR RESP ASPECT ITEM
> > > > > /METHOD = REML
> > > > > /DESIGN = CHAIN SECTOR RESP ASPECT ITEM
> > > > > SECTOR*RESP SECTOR*ASPECT SECTOR*ITEM CHAIN*RESP
> > > > > CHAIN*ASPECT CHAIN*ITEM RESP*ASPECT RESP*ITEM
> > > > > SECTOR*RESP*ASPECT SECTOR*RESP*ITEM
> > > CHAIN*RESP*ASPECT
> > > > > /INTERCEPT = INCLUDE.
> > > > >
> > > > > Thank you for your help.
> > > > >
> > > > > Best regards,
> > > > >
> > > > > Iuri.
> > > > >
> > > > > _______________________________________
> > > > > Iuri Gavronski - iuri@ufrgs.br
> > > > > doutorando
> > > > > UFRGS/PPGA/NITEC - www.ppga.ufrgs.br Brazil
> > > > >
> > > > > ______________________________________________
> > > > > R-help@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
> > > > > and provide commented, minimal, self-contained, reproducible code.
> > > > >
> > > >
> > >
> > > [[alternative HTML version deleted]]
> > >
> > > ______________________________________________
> > > R-help@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
> > > and provide commented, minimal, self-contained, reproducible code.
> > >
> >
> >
> >
>

        [[alternative HTML version deleted]]



R-help@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 and provide commented, minimal, self-contained, reproducible code. Received on Sun Aug 20 23:32:53 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Mon 21 Aug 2006 - 00:22:48 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.