Hi Rui, Data as follows
structure(list(dn = c(4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4), obs = c(1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5, 6, 6, 6, 6, 6, 6, 7, 7, 7, 7, 7, 7, 8, 8, 8, 8, 8, 8, 9, 9), choice = c(0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0), br = c(1, 2, 3, 4, 5, 6, 1, 2, 3, 4, 5, 6, 1, 2, 3, 4, 5, 6, 1, 2, 3, 4, 5, 6, 1, 2, 3, 4, 5, 6, 1, 2, 3, 4, 5, 6, 1, 2, 3, 4, 5, 6, 1, 2, 3, 4, 5, 6, 1, 2), mth = c(NA, NA, NA, NA, NA, 487, NA, NA, 488, NA, NA, NA, NA, NA, NA, NA, NA, 488, NA, NA, 489, NA, NA, NA, NA, NA, NA, NA, NA, 489, NA, NA, NA, NA, NA, 489, NA, NA, NA, NA, NA, 490, NA, NA, NA, NA, NA, 491, NA, NA)), .Names = c("dn", "obs", "choice", "br", "mth"), row.names = c("1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16", "17", "18", "19", "20", "21", "22", "23", "24", "25", "26", "27", "28", "29", "30", "31", "32", "33", "34", "35", "36", "37", "38", "39", "40", "41", "42", "43", "44", "45", "46", "47", "48", "49", "50"), class = "data.frame") Best wishes Graham -----Original Message----- From: Rui Barradas [mailto:ruipbarra...@sapo.pt] Sent: 06 April 2013 16:32 To: Leask, Graham Cc: r-help@r-project.org Subject: Re: [R] Replace missing value within group with non-missing value Hello, Can't you post a data example? If your dataset is named 'dat' use dput(head(dat, 50)) # paste the output of this in a post Rui Barradas Em 06-04-2013 15:34, Leask, Graham escreveu: > Hi Rui, > > Thank you for your suggestion which is very much appreciated. Unfortunately > running this code produces the following error. > > error in '$<-.data.frame' ('*tmp*', "mth", value = NA_real_) : > replacement has 1 rows, data has 0 > > I'm sure there must be an elegant solution to this problem? > > Best wishes > > > > Graham > > On 6 Apr 2013, at 12:15, "Rui Barradas" <ruipbarra...@sapo.pt> wrote: > >> Hello, >> >> That's not a very good way of posting your data, preferably paste the output >> of ?dput in a post. >> Some thing along the lines of the following might do what you want. >> It seems that the groups are established by 'dn' and 'obs' numbers. >> If so, try >> >> >> # Make up some data >> dat <- data.frame(dn = 4, obs = rep(1:5, each = 6), mth = NA) >> dat$mth[6] <- 487 dat$mth[9] <- 488 dat$mth[18] <- 488 dat$mth[21] <- >> 489 dat$mth[30] <- 489 >> >> >> sp <- split(dat, list(dat$dn, dat$obs)) >> names(sp) <- NULL >> tmp <- lapply(sp, function(x){ >> idx <- which(!is.na(x$mth))[1] >> x$mth <- x$mth[idx] >> x >> }) >> do.call(rbind, tmp) >> >> >> Hope this helps, >> >> Rui Barradas >> >> >> Em 06-04-2013 11:33, Leask, Graham escreveu: >>> Dear List members >>> >>> I have a large dataset organised in choice groups see sample below >>> >>> >>> +-------------------------------------------------------------------------------------------------+ >>> | dn obs choice acid br date cdate >>> situat~n mth year set | >>> >>> |-------------------------------------------------------------------------------------------------| >>> 1. | 4 1 0 LOSEC 1 . . >>> . . 1 | >>> 2. | 4 1 0 NEXIUM 2 . . >>> . . 1 | >>> 3. | 4 1 0 PARIET 3 . . >>> . . 1 | >>> 4. | 4 1 0 PROTIUM 4 . . >>> . . 1 | >>> 5. | 4 1 0 ZANTAC 5 . . >>> . . 1 | >>> >>> |-------------------------------------------------------------------------------------------------| >>> 6. | 4 1 1 ZOTON 6 23aug2000 01:00:00 23aug2000 >>> NS 487 2000 1 | >>> 7. | 4 2 0 LOSEC 1 . . >>> . . 2 | >>> 8. | 4 2 0 NEXIUM 2 . . >>> . . 2 | >>> 9. | 4 2 1 PARIET 3 25sep2000 01:00:00 25sep2000 >>> L 488 2000 2 | >>> 10. | 4 2 0 PROTIUM 4 . . >>> . . 2 | >>> >>> |-------------------------------------------------------------------------------------------------| >>> 11. | 4 2 0 ZANTAC 5 . . >>> . . 2 | >>> 12. | 4 2 0 ZOTON 6 . . >>> . . 2 | >>> 13. | 4 3 0 LOSEC 1 . . >>> . . 3 | >>> 14. | 4 3 0 NEXIUM 2 . . >>> . . 3 | >>> 15. | 4 3 0 PARIET 3 . . >>> . . 3 | >>> >>> |-------------------------------------------------------------------------------------------------| >>> 16. | 4 3 0 PROTIUM 4 . . >>> . . 3 | >>> 17. | 4 3 0 ZANTAC 5 . . >>> . . 3 | >>> 18. | 4 3 1 ZOTON 6 20sep2000 00:00:00 20sep2000 >>> R 488 2000 3 | >>> 19. | 4 4 0 LOSEC 1 . . >>> . . 4 | >>> 20. | 4 4 0 NEXIUM 2 . . >>> . . 4 | >>> >>> |-------------------------------------------------------------------------------------------------| >>> 21. | 4 4 1 PARIET 3 27oct2000 00:00:00 27oct2000 >>> NL 489 2000 4 | >>> 22. | 4 4 0 PROTIUM 4 . . >>> . . 4 | >>> 23. | 4 4 0 ZANTAC 5 . . >>> . . 4 | >>> 24. | 4 4 0 ZOTON 6 . . >>> . . 4 | >>> 25. | 4 5 0 LOSEC 1 . . >>> . . 5 | >>> >>> |-------------------------------------------------------------------------------------------------| >>> 26. | 4 5 0 NEXIUM 2 . . >>> . . 5 | >>> 27. | 4 5 0 PARIET 3 . . >>> . . 5 | >>> 28. | 4 5 0 PROTIUM 4 . . >>> . . 5 | >>> 29. | 4 5 0 ZANTAC 5 . . >>> . . 5 | >>> 30. | 4 5 1 ZOTON 6 23oct2000 03:00:00 23oct2000 >>> NS 489 2000 5 | >>> >>> I wish to fill in the missing values in each choice set - delineated by dn >>> (Doctor) obs (Observation number) and choices (1 to 6). >>> For each choice set one choice is chosen which contains full time >>> information for that choice set ie in set 1 choice 6 was chosen and shows >>> the month 487. The other 5 choices show mth as missing. I want to fill >>> these with the correct mth. >>> >>> I am sure there must be an elegant way to do this in R? >>> >>> >>> Best wishes >>> >>> >>> >>> Graham >>> >>> >>> [[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. >> > ______________________________________________ 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.