Hi [EMAIL PROTECTED] napsal dne 04.06.2008 10:57:26:
> I'm new to R so please forgive the newbie question; but i can't seem to > find a definitive answer to this. > > I am wanting to do a PLS regression on some data. Which takes a formula > of the type > "responseACC ~ dataACC" > > where dataACC is multivariate in nature and responseACC is a single value. Really? From which package? AFAIK in package pls you have to enter formula plsr(response~terms, data.frame) > > I have imported my data from a csv file into a dataframe called for > arguments sake df. > This file has no headers in it so I get column names X1.....X83 > > I then create a new data frame which says > data.frame(dataACC = df[1:82], responseACC = df[83]) You probably want data.frame(dataACC = df[,1:82], responseACC = df[,83]) > > When I look at the names of this though I get... > > dataACC.X1 ..... dataACC.X82 and then X83 > > When I pass this to the PLS algorithm I get the error > > "variable responseACC not found" > > So a few questions, Am I doing this in the right way in general? why is > the responseACC not being associated as the name of df[83]? > > Thirdly, If I want to generate my own array of values in order to do get > a predicted response based on my initial PLS regression, will it matter > that the array data will not have header values (eg dataACC.X1 etc)? if > so what the best way of appending the header data? see ?names Regards Petr > > Thanks for any help you can lend > > Chris > > ______________________________________________ > 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.