[R] predict () for LDA and GLM

[Uwe Ligges]

On 22.03.2012 03:24, palanski wrote:
> Hi!
>
> I'm using GLM, LDA and NaiveBayes for binomial classification. My training
> set is 70 rows long with 32 features, and my test set is 30 rows long with
> 32 features.
>
> Using Naive Bayes, I can train a model, and then predict the test set with
> it like so:
>
> ass4q1.dLDA = lda(ass4q1.trainSet[,1]~ass4q1.trainSet[,2:3])
> table(predict(ass4q1.dNB, ass4q1.testSetDF[,2:3]), ass4q1.testSetDF[,1])
>
>
> However, when the same is done for LDA or GLM, suddenly it tells me that the
> number of rows doesn't match and doesn't predict my test data. The error for
> GLM, as an example, is:
>
> Error in table(predict(ass4q1.dGLM, ass4q1.testSetDF[, 2:3]),
> ass4q1.testSetDF[,  :
>    all arguments must have the same length
> In addition: Warning message:
> 'newdata' had 30 rows but variable(s) found have 70 rows
 >
>
> What am I missing?

A correct formula describing the model with separate variables with the 
data.frame passed to the data argument of the lda() function.
A reproducible example is missing, hence this is just a guess.

Uwe Ligges



>
> --
> View this message in context: http://r.789695.n4.nabble.com/predict-for-LDA-and-GLM-tp4494381p4494381.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.