[R] qda problem

[Uwe Ligges]

Christian,

I have reported a similar problem to Brian a few weeks ago. I think it 
has been fixed in recent versions of MASS (in the VR bundle) and works 
for me with another data.frame.

Run update.packages() and try again.

I owed you an answer (you know why),
Uwe


Christian Hennig wrote:

> Hi,
> 
> the following strange error appears when I use qda:
> 
> 
>>qda1 <- qda(as.data.frame(mfilters[cvtrain,]),as.factor(traingroups))
> 
> Error: function is not a closure
> 
> That's also strange:
> 
>>qda1 <- qda(mfilters[cvtrain,],as.factor(traingroups))
> 
> Error in qda.default(mfilters[cvtrain, ], as.factor(traingroups)) : 
> 	length of dimnames must match that of dims
> 
> Some backgroud:
> 
>>str(mfilters[cvtrain,])
> 
>  num [1:12500, 1:12] -0.426  0.937 -1.610 -2.099  0.749 ...
>  - attr(*, "dimnames")=List of 2
>   ..$ : chr [1:12500] "1" "2" "8" "9" ...
>   ..$ : NULL
> 
>>str(as.factor(traingroups))
> 
>  Factor w/ 3 levels "1","2","3": 1 1 1 1 1 1 1 1 1 1 ...
>  - attr(*, "names")= chr [1:12500] "1" "2" "8" "9" ...
> 
>>str(as.data.frame(mfilters[cvtrain,]))
> 
> `data.frame':	12500 obs. of  12 variables:
>  $ V1 : num  -0.426  0.937 -1.610 -2.099  0.749 ...
>  $ V2 : num   0.7970 -1.9004  0.0443 -1.2074  0.4095 ...
>  $ V3 : num  -0.303 -0.636 -0.806  0.639  0.363 ...
>  $ V4 : num   0.130 -0.096 -0.644  0.723  0.576 ...
>  $ V5 : num  -0.2942  1.4951 -0.0098  0.3253  0.3881 ...
>  $ V6 : num  -0.2943 -0.6561 -0.0863 -0.0849 -0.5070 ...
>  $ V7 : num   0.512 -0.618 -0.244  0.392  0.346 ...
>  $ V8 : num  -0.171  0.677  0.117 -0.113  0.669 ...
>  $ V9 : num   0.2289 -0.3934  0.1051  0.1545 -0.0446 ...
>  $ V10: num   0.0188  0.5614 -0.2271  0.0340  0.2207 ...
>  $ V11: num  -0.152  0.631  0.447  0.696  0.458 ...
>  $ V12: num   0.6139 -0.9379 -1.1784  0.0802 -0.6625 ...
> 
> ...looks proper to me and works without problems with svm...
> 
> Even stranger is the fact that my .R file suggests that the first command
> qda1 <- qda(as.data.frame(mfilters[cvtrain,]),as.factor(traingroups))
> worked in December for the same data (apart from random sampling of 
> cvtrain/traingroups, but I tried more than one version), and
> even under the same version of R (1.8.0).
> 
> Can anybody tell me what goes wrong now?
> 
> Best,
> Christian
> 
> ***********************************************************************
> Christian Hennig
> Fachbereich Mathematik-SPST/ZMS, Universitaet Hamburg
> hennig at math.uni-hamburg.de, http://www.math.uni-hamburg.de/home/hennig/
> #######################################################################
> ich empfehle www.boag-online.de
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html