Your 'newdata' is a vector. To us predict() the newdata supplied must be a data frame, with components having the same names as the predictors used in the model. If some of the variables are factors, then the corresponding factors in the newdata data frame must have the same names and the same levels as the original factors used in fitting the model *even if* not all levels are needed for the prediction.
This is a precise sort of area. You may need to study the help information in a bit more detail. predict() does work, if only you use it correctly. Bill Venables http://www.cmis.csiro.au/bill.venables/ -----Original Message----- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of serbring Sent: Wednesday, 29 July 2009 11:31 PM To: r-help@r-project.org Subject: [R] predict I have found a regression model, and i would like to predict value in different points. I have tried to use predict function but it doesn't work. I have used predict function like this: newdata<-seq(from=0.1, to=0.32,by=0.02) data<-predict(fm,newdata) where fm is a regression model. The predict function return me that: Error in eval(predvars, data, env) : numeric argument 'envir' doesn't have unitary length where is the error? -- View this message in context: http://www.nabble.com/predict-tp24719362p24719362.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help@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. ______________________________________________ R-help@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.