[R] tapply and more than one function, with different arguments

[RINNER Heinrich]

Hi Dennis,
now that's a very nice function, and this seems to be just what I need!
Thanks a lot!
-Heinrich.
________________________________________
Von: Dennis Murphy [djmuser at gmail.com]
Gesendet: Dienstag, 26. Januar 2010 19:44
An: RINNER Heinrich
Cc: r-help
Betreff: Re: [R] tapply and more than one function, with different arguments

Hi:

Using the plyr package, we can get the result as follows:

> library(plyr)
> my.fun <- function(x, mult) mult*sum(x)
> dat <- data.frame(x = 1:4, grp = c("a","a","b","b"))
> ddply(dat, .(grp), summarize, max = max(x), myfun = my.fun(x, 10))
  grp max myfun
1   a   2    30
2   b   4    70

HTH,
Dennis

On Tue, Jan 26, 2010 at 8:26 AM, RINNER Heinrich <HEINRICH.RINNER at tirol.gv.at<mailto:HEINRICH.RINNER at tirol.gv.at>> wrote:
Dear R-users,

I am working with R version 2.10.1.

Say I have is a simple function like this:

> my.fun <- function(x, mult) mult*sum(x)

Now, I want to apply this function along with some other (say 'max') to a simple data.frame, like:

> dat <- data.frame(x = 1:4, grp = c("a","a","b","b"))

Ideally, the result would look something like this (if mult = 10):
 max my.fun
a   2     30
b   4     70

I have tried it that way:

apply.more.functions <- function(dat, FUN = c("max", "my.fun"), ...) {
 res <- NULL
 for(f in FUN) res[[f]] <- tapply(dat$x, dat$grp, FUN = f, ...)
 data.frame(res)
}

# let's test it:
> apply.more.functions(dat, FUN = c("max", "min"))
 max min
a   2   1
b   4   3
# perfect!

# now, with an additional argument:
> apply.more.functions(dat, FUN = c("max", "my.fun"), mult = 10)
 max my.fun
a  10     30
b  10     70
# uhuh!
Apparently, 'mult' has been used in the calculation of 'max' as well.
How can I modify apply.more.functions in order to avoid this?

Your advice would be appreciated;
Kind regards
Heinrich.

______________________________________________
R-help at r-project.org<mailto:R-help at 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.