[R] Recode variable as NA

[Prof Brian Ripley]

On Sat, 30 Oct 2004, Neil Leonard wrote:

> I was wondering how I would recode a variable as being NA (or missing).
It is the values, not the variable that are NA.  For example

x <- 1:20
ind <- x%%4==0
ind
x[ind] <- NA  # or is.na(x[ind]) <- TRUE

That works for logical, integer, numeric, complex and character variables.
(Raw and list vecctors do not have missing values.)

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595