On Mar 23, 2011, at 10:05 AM, agent dunham wrote:
> Dear all,
>
> I've fitted a lm using 61 data (training data), and I'left 10 as
> test data.
>
> Training data and test data are stored in an excell.
>
> training <- read.xls("C:/...../training.xls") , the same for
test.
> That is:
> v1
> v2
> ...
> v15
>
> When I type str(training) and str(test), both sets have the same names
>
> The resulting model is lms <- lm(vd ~ log(v1) + fv2+ fv5+ fv7 ) -
> fvi means
> they were turned into factors-
>
> plms<- predict(lms, new=test ,interval="prediction")
>
> Error at model.frame.default(Terms, newdata, na.action = na.action,
> xlev > object$xlevels) :
> length of the variables are different (found for 'fv2')
> More: Warning messages lost
> 'newdata' had 10 rows but variable(s) found have 61 rows
>
> q1: What does it mean?
In the absence of a reproducible example it is difficult to say.
> q2: Do I have to change test data names, so they have the same as the
> resulting lm?
At a minimum, that would be required. Read help(predict.lm) and pay
particular attention to _all_ the details mentioned in the newdata
section.
> q3: Do I have to do anything special because of the log
> transformation?
No.
> q4: Afterwards I'd like to plot it, is this way?: plot(plms)
Specify what kind of plot. You are likely to be surprised and perhaps
disappointed if you use the default plot method for lm objects.
As the standard postscript says:> and provide commented, minimal, self-contained, reproducible code.
--
David Winsemius, MD
West Hartford, CT