[R] Problems with stepAIC
Katja
katja.hoelzer at onlinehome.de
Sat Sep 29 09:36:24 CEST 2012
Dear help community,
I'm a R-beginner and use it for my master thesis.
I've got a mixed model and want to analyse it with lme. There are a lot
Cofactors that coult be relevant. To extract the important ones I want to do
the stepAIC, but always get an error warning.
Structure of my data:
data.frame': 72 obs. of 54 variables:
$ Block : Factor w/ 3 levels "A","B","C": 1 1 1 1 1 1
1 1 1 1 ...
$ Treatment : Factor w/ 3 levels "1","2","3": 1 1 1 1 1 1
1 1 2 2 ...
$ Paddock : Factor w/ 9 levels "A1","A2","A3",..: 1 1 1
1 1 1 1 1 2 2 ...
$ Cluster : Factor w/ 4 levels "1","2","3","4": 1 1 2 2
3 3 4 4 1 1 ...
$ Growth : Factor w/ 2 levels "s","t": 1 2 1 2 1 2 1 2
1 2 ...
$ Patch : Factor w/ 72 levels "A1_1s","A1_1t",..: 1 2
3 4 5 6 7 8 9 10 ...
$ Lufttemperatur : num 17.3 17.3 17.3 17.3 17.3 ...
$ NSmm : num 0 0 0 0 0 0 0 0 0 0 ...
$ Luftfeuchte : num 79.2 79.2 79.2 79.2 79.2 ...
$ Windstärke : num 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96
1.27 1.27 ...
$ Bodentemperatur : num 18.1 18.1 18.1 18.1 18.1 ...
$ Earthwormsm.2 : num 0 0 0 20 24 60 12 20 0 4 ...
$ Biomassgm.2 : num 0 0 0 24.5 17.8 ...
$ meanBiomassg : num 0 0 0 1.227 0.743 ...
$ Species : num 0 0 0 1 2 2 0 0 0 1 ...
$ Juvenilem.2 : num 0 0 0 4 4 1 2 3 0 0 ...
$ JuvenileRate : num NA NA NA 80 80 ...
$ AdultsRate : num NA NA NA 20 20 ...
$ Pflanzenfrischmasse : num 139 803 189 739 261 ...
$ Pflanzentrockenmasseanteil : num 27.8 29.3 28.7 40 27.7 ...
$ NAnteilGesamt : num 2.38 3.35 2.47 2.21 2.55 4.45 2.33 2.23
2.3 1.68 ...
$ CAnteilGesamt : num 43.9 45 45 43.6 43.8 ...
$ aBodenfeuchte : num 24.4 21.8 14.7 18.2 23.9 ...
$ bBodenfeuchte : num 17.3 15.6 NA NA 19.1 ...
$ cBodenfeuchte : num NA NA NA NA NA ...
$ Gfrischmasse : num 84.6 739.3 130.8 566.8 94 ...
$ Gtrockenmasse : num 25.6 215.4 41.7 239.4 30.5 ...
$ Gtrockenmasseanteil : num 30.2 29.1 31.9 42.2 32.5 ...
$ GtrockenmasseanteilanGesamt: num 78.8 98.3 86 96 76.5 ...
$ GN : num 2.1 3.35 2.31 2.17 2.28 ...
$ GC : num 43.8 45 44.9 43.6 43.5 ...
$ GCN : num 20.9 13.4 19.4 20.1 19.1 ...
$ Lfrischmasse : num 12.1 0 0 0 19 ...
$ Ltrockenmasse : num 2.56 0 0 0 3.92 1.84 0 0 2.2 0 ...
$ Ltrockenmasseanteil : num 21.2 0 0 0 20.7 ...
$ LtrockenmasseanteilanGesamt: num 7.89 0 0 0 9.83 1.06 0 0 3.73 0 ...
$ LN : num 3.96 NA NA NA 4.09 ...
$ LC : num 43.1 NA NA NA 45.6 ...
$ LCN : num 10.9 NA NA NA 11.1 ...
$ Kfrischmasse : num 20.16 8.96 38.48 56.24 31.2 ...
$ Ktrockenmasse : num 4.32 3.64 6.8 10 5.44 ...
$ Ktrockenmasseanteil : num 21.4 40.6 17.7 17.8 17.4 ...
$ KTrockenmasseanteilanGesamt: num 13.32 1.66 14.01 4.01 13.64 ...
$ KN : num 3.09 3.11 3.43 3.37 2.95 ...
$ KC : num 45 48.3 46 44 44.3 ...
$ KCN : num 14.6 15.5 13.4 13.1 15 ...
$ I : num 2.12 1.4 0.92 0.49 1.03 0.53 1.36 0.51
1.1 2.31 ...
$ II : num 2.61 3.97 2.82 NA 1.94 1.98 2.84 1.66
2.38 1.99 ...
$ III : num 2.24 3.51 3.17 NA 2.95 3.12 2.16 3.65
3.14 2.83 ...
$ IV : num 2.52 2.91 3.11 NA 1.97 4.39 3.22 4.61
2.26 2.77 ...
$ V : num 2.38 2.91 2.61 NA 3.34 3.32 3.21 NA
2.44 2.26 ...
$ VI : num 2.01 2.61 2.61 NA 3.61 3.57 NA NA 3.06
2.18 ...
$ VII : num NA 3.73 2.61 NA 3.25 0.27 NA NA NA 3.04
...
$ Eindringtiefe : int 64 80 80 8 80 61 55 32 73 80 ...
model fo the first parameter I want to check:
PModell1 <-lme(sqrt(Earthwormsm.2)~
Treatment+Pflanzenfrischmasse+aBodenfeuchte+bBodenfeuchte+Gfrischmasse+Ltrockenmasseanteil+KCN+I+Eindringtiefe,
random=~1|Block/Treatment/Cluster/Patch, data=Test1,na.action = na.omit)
Everytime I try stepAIC with different settings:
>Modell2<- stepAIC(PModell1, method=lme)
>Fehler in extractAIC.lme(fit, scale, k = k, ...) :
AIC für REML-Näherung undefiniert
englisch: Error in extractAIC.lme(fit, scale, k = k, ...) : AIC for
REML-Approximation undefined
Same when I do extractAIC:
>extractAIC(PModell1)
>Fehler in extractAIC.lme(PModell1) : AIC undefined for REML fit
Can anybody tell me what's wrong here? Package MASS is running, there ist no
error.
--
View this message in context: http://r.789695.n4.nabble.com/Problems-with-stepAIC-tp4644583.html
Sent from the R help mailing list archive at Nabble.com.
More information about the R-help
mailing list