Re: [R] replacing zeros by NAs

Often, data are computed in real (not integer) numbers, and in such cases, comparison to zero is not as good as using a tolerance level. Modifying slightly the suggestions of others, I would use something like this: my.tol <- 1e-50 x <- c(1.4, 2.97, 1.0e-100, pi, 0.2, 5.1, 6.8, 0.0) print(x) is.na

Re: [R] replacing zeros by NAs

Try this also: is.na(x) <- x == 0 On Fri, Jun 12, 2009 at 6:59 AM, Robert Kinley wrote: > something like ... > > > x<-c(1,2,3,0,5,6,0) > > is.na(x[x==0])<-T > > x > [1] 1 2 3 NA 5 6 NA > > > Robert Kinley > > > > >[[alternative HTML version delete

Re: [R] replacing zeros by NAs

Other method: a<-matrix(seq(1,20,2), 4,5) a[c(1:2),c(2:4)]<-0 a[a==0]<-NA - Original Message - From: "Robert Kinley" To: Sent: Friday, June 12, 2009 11:59 AM Subject: [R] replacing zeros by NAs something like ... x<-c(1,2,3,0,5,6,0) is.na(x[x==0])<-T

[R] replacing zeros by NAs

something like ... > x<-c(1,2,3,0,5,6,0) > is.na(x[x==0])<-T > x [1] 1 2 3 NA 5 6 NA Robert Kinley [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailma