[R] Running stats on individual data.frames from the split()function list

[Daniel Malter]

This easy function you are looking for is tapply. Take a look at the
following example:

day=rep(1:30,each=30)
##There are thirty days
##with thirty obs each

y=rnorm(length(day),mean=2*day,sd=day)
##a dep. variable
##with mean=2*day index no.
##and sd=day

tapply(y,day,length)
##shows no. of obs for each day

tapply(y,day,mean)
##shows mean of y for each day
##should be about equal to two times the day index no.

tapply(y,day,sd)
##show sd of y for each day
##should be about equal to the day index no. 

Best,
Daniel

-------------------------
cuncta stricte discussurus
-------------------------

-----Ursprüngliche Nachricht-----
Von: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] Im
Auftrag von Eric Vander Wal
Gesendet: Tuesday, June 16, 2009 4:44 PM
An: r-help at r-project.org
Betreff: [R] Running stats on individual data.frames from the
split()function list

Hello, and thanks in advance.

I have a data.frame from which I want to count observations that occur on
each day and determine the mean and std.error of said counts.

For instance:

x<-split(my.df, my.df$julian.days)

Although I'm still in my R learning infancy I am under the impression that x
is a list of data.frames subsetting my.df by group
(i.e.,julian.day) where day 1:366 are x$'1': x$'366' and my variables are
x$'1'$var1, x$1$var2, for each data.frame in the list.

The data I seek can be supplied by

mean(sapply(split(x$'1'$var1, x$'1'$var2), length)) and
std.error(sapply(split(x$'1'$var1, x$'1'$var2), length)), etc.

Is there an efficient means for me to process the entire list x so that I
needn't call each data.frame (i.e., day) from list x individually?  No doubt
there is a more sophisticated way to obtain this information from the
outset, but I'm not yet familiar with it.

Thanks again for all the help,
Eric

--
Eric Vander Wal
Ph.D. Candidate
University of Saskatchewan, Department of Biology,
112 Science Place, Saskatoon, SK., S7N 5E2

"Pluralitas non est ponenda sine neccesitate"

______________________________________________
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.