[R] Help on R Coding

andyer weng nzandyer at gmail.com
Sun Oct 5 05:25:58 CEST 2008


Hi all,

I am kind of stuck of  using Predict function in R to make prediction
for a model with continuous variable and categorial variables. i have
no problem making the model, the model is e.g.

cabbage.lm2<- lm(VitC ~ HeadWt + Date + Cult)

HeadWt is a continuous variable, Date and Culte are factors. Date have
three levels inside (d16,d20,d21), Cult has two levels(c39,c52). I
need to calculate a confidence interval for the mean VitC for each
combination of Date and Cult, fixing the value of HeadWt at the mean
for the corresponding cell. I have already proved that Cult and Date
are not interacted. the mean of HeadWt is also found. e.g.2.59

when i type

> new.df<-data.frame(HeadWt=2.59,Cultc52=1,Dated16=1)
> predict(cabbage.lm2,new.df, interval="confidence")

it has error comes up like this:
Error in model.frame.default(Terms, newdata, na.action = na.action,
xlev = object$xlevels) :
  variable lengths differ (found for 'Cult')
In addition: Warning message:
'newdata' had 1 rows but variable(s) found have 60 rows


Is there anything I have done wrong?? Please help with the coding.

Thank you so much!!!

All the best.

Andyer.



More information about the R-help mailing list