HI, May be this helps: a1 <-a
indx <- which(c(0,diff(a!=0)) <0) indx1 <- which(c(0,diff(a!=0)) >0) a[a==0] <- rep( rowMeans(cbind(a[indx-1],a[indx1])),indx1-indx) a # [1] 0.97210 0.97220 0.97300 0.97230 0.97145 0.97145 0.97145 0.97060 0.96980 #[10] 0.97040 0.97100 0.96990 #Another option is na.approx() from library(zoo) a1[a1==0] <- NA na.approx(a1) ##there is difference in the result though (depends on what you want) [1] 0.972100 0.972200 0.973000 0.972300 0.971875 0.971450 0.971025 0.970600 [9] 0.969800 0.970400 0.971000 0.969900 A.K. Dear all, My data is : a <- c(0.9721,0.9722,0.9730,0.9723,0.0,0.0,0.0,0.9706,0.9698,0.0,0.9710,0.9699) I want to replace zeros with average of before and after values of them. But sometimes there is one zero sometimes more than one. What is the most elegant way to do this ? Thanks a lot ______________________________________________ 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.
