R - Recoding (missing) values in a matrix

let v be a vector
> v <- c(1,2,4,99,6,999,7,8,99,5,2,4,9,2)

where the missing values where coded as 9, 99 or 999.

Then the fastest way (i know) to recode them is:

> v[v %in% c(9,99,999)] <- NA

> v
[1] 1 2 4 NA 6 NA 7 8 NA 5 2 4 NA 2

an example with matrices:
> m <- matrix(1:12, nrow=3, byrow=T,dimnames = list(letters[1:3],LETTERS[1:4])) > m
A B C D
a 1 2 3 4
b 5 6 7 8
c 9 10 11 12
> m[m %in% c(9,99,999)] <- NA

> m
A B C D
a 1 2 3 4
b 5 6 7 8
c NA 10 11 12

This does not work on dataframes. You have to operate on each column separately. If you have many columns maybe it is worth a try to load the package epicalc or lordif and use the command recode() or you do it in loop ...