Skip to content

Observation-level random effect to model overdispersion

4 messages · M.S.Muller, Ben Bolker, Jarrod Hadfield +1 more

#
Dear all,

I'm trying to analyze some strongly overdispersed Poisson-distributed data using R's mixed effects model function "lmer". Recently, several people have suggested incorporating an observation-level random effect, which would model the excess variation and solve the problem of underestimated standard errors that arises with overdispersed data. It seems to be working, but I feel uneasy using this method because I don't actually understand conceptually what it is doing. Does it package up the extra, non-Poisson variation into a miniature variance component for each data point? But then I don't understand how one ends up with non-zero residuals and why one can't just do this for any analyses (even with normally-distributed data) in which one would like to reduce noise.

I may be way off base here, but does this approach model some kind of mixture distribution that's a combination of Poisson and whatever distribution the extra variation is? I've read that people often use a negative binomial distribution (aka Poisson-gamma) to model overdispersed count data in which they assume that the process is Poisson (so they use a log link) but the extra variation is a gamma distribution (in which variance is proportional to square of the mean). The frequently referred to paper by Elston et al (2001) describes modeling a Poisson-lognormal distribution in which overdispersion arises from errors taking on a lognormal distribution. Is the approach of using the observation-level random effect doing something similar, and simply assuming some kind of Poisson-normal mixed distribution? Does this approach therefore assume that the observation-level variance is normally distributed?

If anyone could give me any guidance on this, I would appreciate it very much.

Martina Muller
#
On 11-03-21 07:51 AM, M.S.Muller wrote:
Exactly.
  The observation-level random effect approach is equivalent to assuming
that the individual observations are [x]-normal distributed, i.e. a
compound of a normal distribution transformed by the inverse link
function and the specified distribution family.

  Sorry that's a bit clunky, but it translates to what you said above --

* lognormal-Poisson for Poisson with log link;
* logit-normal-binomial for binomial with logit link;
  etc.

  For what it's worth, the Elston paper is philosophically sensible but
I'm not sure that it's computationally sound; as I have said before
<https://stat.ethz.ch/pipermail/r-sig-mixed-models/2010q2/003967.html>,
using PQL with observation-level random effects is explicitly
*dis*recommended in the Genstat documentation; I had convergence
problems fitting the data in lme4, and MCMCglmm told me that the data
were under-specified and I should consider a more informative prior ...

  (I see Jarrod Hadfield has just answered this question too.)
#
Hi,

Your intuition is correct: using an observation-level random effect is  
the same as using a log-normal mixing distribution if the log-link is  
used: the marginal distribution of the observation-level effects are  
assumed to be normal. It goes back to HINDE J (1982) GLIM 82-109  at  
least.

Jarrod
On 21 Mar 2011, at 11:51, M.S.Muller wrote:

            
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-mixed-models/attachments/20110321/d4cebe97/attachment.pl>
#
One point, additional to other responses.  There is just one 
"miniature variance component" (by the way, not miniature 
in the sense that it has to be small).  As I understand it, a 
normal distribution with this variance generates one random 
effect, on the scale of the linear predictor, for each observation.  
On the scale of the response, well . . .

John Maindonald             email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473    fax  : +61 2(6125)5549
Centre for Mathematics & Its Applications, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
http://www.maths.anu.edu.au/~johnm
John Maindonald             email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473    fax  : +61 2(6125)5549
Centre for Mathematics & Its Applications, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
http://www.maths.anu.edu.au/~johnm
On 21/03/2011, at 10:51 PM, M.S.Muller wrote: