you should look at the 'na.rm=FALSE' argument of '?mean()', i.e.,
x <- rnorm(100); x[sample(100, 10)] <- NA
f <- sample(letters[1:5], 100, TRUE)
###############
tapply(x, f, mean)
tapply(x, f, mean, na.rm=TRUE)
I hope it helps.
Best,
Dimitris
----
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven
Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/16/336899
Fax: +32/16/337015
Web: http://www.med.kuleuven.ac.be/biostat/
http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm
----- Original Message -----
From: "Leonardo Lami" <lami at faunalia.it>
To: <r-help at stat.math.ethz.ch>
Sent: Friday, March 25, 2005 10:35 AM
Subject: [R] tapply and NA value
Hi,
I'm writing for a little help.
I have a dataframe with same NA value and I'd like to obtain the
means of the
value of a coloumn grouped by the levels of a factor coloumn of the
datframe.
I'm using the function "tapply" but I see that if only a NA value is
present
the result is NA.
There is an option to have the correct result or I must use an other
function?
Thanks of all
Leonardo
--
Leonardo Lami
lami at faunalia.it www.faunalia.it
Via Colombo 3 - 51010 Massa e Cozzile (PT), Italy Tel:
(+39)349-1310164
GPG key @: hkp://wwwkeys.pgp.net http://www.pgp.net/wwwkeys.html
https://www.biglumber.com