It is not entirely clear what type of transformation you are trying to do. Can you provide some sample data and then show what you would expect the output to look like; make sure the data covers all the cases you want to transform.
On Sat, Mar 13, 2010 at 8:54 PM, Junqian Gordon Xu <xjq...@gmail.com> wrote: > I have a multilevel dataframe (df): > > ID Date Segment Slice Tract Lesion > 1 CSPP005 12/4/2007 1 1 LCST 0 > 2 CSPP005 12/4/2007 1 1 LPC 2 > 3 CSPP005 12/4/2007 1 1 RPC 3 > 4 CSPP005 12/4/2007 1 1 RCST 1 > 5 CSPP005 12/4/2007 1 1 LGM 0 > 6 CSPP005 12/4/2007 1 1 RGM 0 > 7 CSPP005 12/4/2007 1 1 Whole NA > > The five levels are: > > ID->Date->Segment->Slice->Tract > > The observation variable is: Lesion > > What I want to do is (in pseudo code) > > if (any of the Tract (for a given ID, Date, Segment, and Slice), > which( Lesion == 2) ) > df [ for that particular (ID, Date,Segment, and slice) , > which(Tract == Whole) ] $Lesion <- 2 > else > df [ for that particular (ID, Date,Segment, and slice) , > which(Tract == Whole) ] $Lesion <- 0 > > > > I started with (don't know if this is the right path), > > Lesion2<-df[which(df$Lesion == 2),] > Where.Lesion2<-unique(Lesion2[,1:4]) > Whole<-subset(df, Tract == "Whole") > > But stuck at how to match the ID/Date/Segment/Slice from Where.Lesion2 to > Whole. > > Regards > Gordon > > ______________________________________________ > 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<http://www.r-project.org/posting-guide.html> > and provide commented, minimal, self-contained, reproducible code. > -- Jim Holtman Cincinnati, OH +1 513 646 9390 What is the problem that you are trying to solve? [[alternative HTML version deleted]] ______________________________________________ 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.
