[R-sig-ME] No data for 1 interaction combination: problem in R not in Genstat

[ONKELINX, Thierry]

Dear Stephanie,

IMHO the problem is that your model is too complex for your data. The model needs 14 d.f. and your data has only 22 observations.

Best regards,

Thierry

----------------------------------------------------------------------------
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek
team Biometrie & Kwaliteitszorg
Gaverstraat 4
9500 Geraardsbergen
Belgium

Research Institute for Nature and Forest
team Biometrics & Quality Assurance
Gaverstraat 4
9500 Geraardsbergen
Belgium

tel. + 32 54/436 185
Thierry.Onkelinx at inbo.be
www.inbo.be

To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of.
~ Sir Ronald Aylmer Fisher

The plural of anecdote is not data.
~ Roger Brinner

The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey


> -----Oorspronkelijk bericht-----
> Van: r-sig-mixed-models-bounces at r-project.org [mailto:r-sig-mixed-models-
> bounces at r-project.org] Namens stéphanie braun
> Verzonden: dinsdag 8 november 2011 19:33
> Aan: r-sig-mixed-models at r-project.org
> Onderwerp: [R-sig-ME] No data for 1 interaction combination: problem in R not
> in Genstat
> 
> Dear list member
> I have a similar problem like Roger, but more complex...
> I have two interaction terms that are unoccupied and one interaction term that
> is complete. I would like to test all in the same linear mixed-effects model.
> 
> my data is the following:
> 
>  str(mydata)
> 'data.frame':   22 obs. of  12 variables:
>  $ hare         : chr  "AF" "AH" "AH" "AH" ...
>  $ period       : num  1 2 3 4 5 15 17 6 16 18 ...
>  $ mcp          : num  11.22 4.6 6 8.11 10.51 ...
>  $ k_href       : num  19.25 6.21 7.94 12.14 11.61 ...
>  $ k_hnew       : num  18.36 7.89 9.83 9.84 12.57 ...
>  $ activity     : chr  "a-na" "a-na" "a-na" "a-na" ...
>  $ month        : num  5 7 7 7 7 7 7 7 7 7 ...
>  $ sexfactor    : Factor w/ 2 levels "f","m": 2 1 1 1 1 1 2 1 1 1 ...
>  $ monthfactor  : Factor w/ 4 levels "5","7","8","9": 1 2 2 2 2 2 2 2 2 2 ...
>  $ harvestfactor: Factor w/ 2 levels "harvest","post_harvest": 1 1 1 1 1 1 1 2 2 2
> ...
>  $ monthsex     : Factor w/ 6 levels "f:7","f:8","f:9",..: 4 1 1 1 1 1 5 1 1 1 ...
>  $ monthharvest : Factor w/ 5 levels "harvest:5","harvest:7",..: 1 2 2 2 2 2 2 3 3 3
> ...
> 
> I would like to test the effect of harvest, sex and season (month) on the home-
> range size (k_hnew) of hares.
> 
> monthsex and monthharvest are the interaction terms with unoccupied levels.
> Another interaction term I would like to include in my analysis is sex*harvest (the
> random factors are week and hare).
> I tried the following mixed-effects model:
> 
> model_h_r<-
> lmer(sqrt(k_hnew)~factor(sex)*factor(harvest)+monthsex+monthharvest+(1|we
> ek)+(1|hare),data=mydata)
> object=pvals.fnc(model_h_r, nsim = 10000) object$fixed
> 
> I get the error message:
> error in mer_finalize(ans) : Downdated X'X is not positive definite, 8.
> 
> If I include only one interaction term in the linear mixed-effects model , I don't
> get any error message!
> 
> model_h_r_monthsex<-
> lmer(sqrt(k_hnew)~monthsex+(1|week)+(1|hare),data=mydata)
> object=pvals.fnc(model_h_r_monthsex, nsim = 10000) object$fixed
> 
> model_h_r_harvest_week<-
> lmer(sqrt(k_hnew)~factor(sex)*factor(harvest)+(1|week)+(1|hare),data=mydata
> )
> object=pvals.fnc(model_h_r_harvest_week, nsim = 10000) object$fixed
> 
> model_h_r_monthharvest<-
> lmer(sqrt(k_hnew)~monthharvest+(1|week)+(1|hare),data=mydata)
> object=pvals.fnc(model_h_r_monthharvest, nsim = 10000) object$fixed
> 
> I woul very much appreciate your help!
> Thank you,
> Stéphanie