Skip to content

Crossed random effects

7 messages · Douglas Bates, Martin Henry H. Stevens, Kevin Wright +3 more

#
I am confused by some apparent contradictions about fitting crossed
random effects in software.  Consider this quote from
http://www.mpi.nl/world/persons/private/baayen/publications/baayenDavidsonBates.pdf
"To our knowledge, the only software currently available for fitting
mixed-effects models with crossed random effects is the lme4 package"

Yet, nlme and GLIMMIX appear to claim that crossed-random effects can
be fit by those respective tools:

In Mixed Effects Models in S and S-Plus:
"The crossed random-effects structure is represented in lme by a
combination of pdBlocke3d and pdIdent objects" (page 163)

http://support.sas.com/rnd/app/papers/glimmix.pdf
"The GLIMMIX procedure, on the other hand, determines by default the
marginal log likelihood as that of an approximate linear mixed model.
This allows multiple random effects, nested and crossed random
effects, multiple cluster types, and R-side random components."  [and]
 "Example 2. Mating Experiment with Crossed Random Effects"

Are these three quotes using different definitions of "crossed random
effects"?  Have I taken the quotes out of context?  Any clarifications
would be appreciated.

Thanks,

K Wright
#
On 3/13/07, Kevin Wright <kw.statr at gmail.com> wrote:
That statement should have been more carefully worded.  It is in
reference to the types of experimental situations described in that
paper where random effects are associated with subject and item,
subjects are crossed with item and the numbers of both the subjects
and the items can be very large.
It is possible to fit a model with crossed random effects with lme
provided that the number of levels of both of the crossed factors is
small.  Otherwise you end up with huge, sparse model matrices that are
being treated as dense matrices and you quickly run out of memory or
time or both.

Really, doesn't a random effects specification like
pdBlocked(list(pdIdent(~ rows - 1), pdIdent(~ columns - 1))) smell
like a kludge to you?
I think that several readers of this list could tell you war stories
of trying to fit models with crossed random effects using SAS PROC
MIXED or SAS PROC NLMIXED versus fitting the same model in lmer or
lmer2.  You are correct that one can specify a model with crossed
random effects in SAS PROC MIXED and that we overstated the uniqueness
of the capabilities of lmer to fit such models.  However, if you want
to try to fit such a model in SAS PROC MIXED when you have large
numbers of subjects and large numbers of items you had better be
prepared to wait for a long time.
#
Thanks for the clarification.  It is no secret that large
plant-breeding programs (both corporate and governmental--see
http://www.dpi.nsw.gov.au/__data/assets/pdf_file/113474/annual_report_part_3.pdf)
have adopted ASREML, probably due to the "war stories" with crossed
random effects that you mention.  I have heard several people say that
ASREML is often orders of magnitude (100-1000) times better than SAS
for handling large datasets with crossed random effects.  My limited
experience suggests ASREML/Genstat/SAMM and lme4 are in the same
order-of-magnitude performance-wise.

P.S.  I offer sincere appreciation for the "Mixed-effects modeling
with crossed random effects for subjects and items" paper,
particularly the MCMC approach and the corresponding interpretations
and discussions.  Very nice.

K Wright
On 3/13/07, Douglas Bates <bates at stat.wisc.edu> wrote:
#
On Tue, Mar 13, 2007 at 02:31:56PM -0500, Douglas Bates wrote:
You must have one of those fancy new monitors.

Cheers,

Andrew
#
Do try glmm.admb() from library(glmmADMB)in R. It is in the development
phase with limited functionality for now. However, the results are very
fast with large data sets. One can try the base software ADMB
implemented in C which is claimed to be very fast.

Monica

-----Original Message-----
From: r-sig-mixed-models-bounces at r-project.org
[mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of Doran,
Harold
Sent: Tuesday, March 13, 2007 3:50 PM
To: MHH Stevens; Douglas Bates
Cc: r-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] Crossed random effects

I've missed some prior threads on this, please accept my apologies if
what I say below has already been noted. It is true that lme in the nlme
package and that HLM (stand-alone) can fit models with crossed random
effects. Now, I just dare you to try it. mlWin uses an MCMC
implementation for crossed random effects (if you want to go down that
road). 

I have some recent experiences fitting models in Stata and in R. Models
that took less than 2 minutes in R would take overnight in Stata. A few
years back, I also did some comparisons with HLM. For a small data set,
a model in lmer that could be fit in less than 1 minute took something
like 3 to 4 hours in HLM.

In the JSS special edition on psychometrics (forthcoming) Doug, Paul
Bliese, Maritza dowling and I estimate the 1PL for items and students
that are fully crossed using lmer. The estimates were resolved extremely
fast and the data set was rather large.  

I have simply not found another package that competes with lmer wrt to
computational speed for linear or generalized linear mixed models.

Harold



-----Original Message-----
From: r-sig-mixed-models-bounces at r-project.org on behalf of MHH Stevens
Sent: Tue 3/13/2007 5:14 PM
To: Douglas Bates
Cc: r-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] Crossed random effects
 
Dear Folks,
What about specialized stand alone mixed model software, such as HLM?
-Hank
On Mar 13, 2007, at 3:31 PM, Douglas Bates wrote:

            
Dr. Hank Stevens, Assistant Professor
338 Pearson Hall
Botany Department
Miami University
Oxford, OH 45056

Office: (513) 529-4206
Lab: (513) 529-4262
FAX: (513) 529-4243
http://www.cas.muohio.edu/~stevenmh/
http://www.muohio.edu/ecology/
http://www.muohio.edu/botany/

"If the stars should appear one night in a thousand years, how would men
believe and adore." -Ralph Waldo Emerson, writer and philosopher  
(1803-1882)







_______________________________________________
R-sig-mixed-models at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models



_______________________________________________
R-sig-mixed-models at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models


#########################################################
The information contained in this e-mail and subsequent attachments may be privileged, 
confidential and protected from disclosure.  This transmission is intended for the sole 
use of the individual and entity to whom it is addressed.  If you are not the intended 
recipient, any dissemination, distribution or copying is strictly prohibited.  If you 
think that you have received this message in error, please e-mail the sender at the above 
e-mail address.
#########################################################