[R] x y averaging

[Prof Brian Ripley]

It's confusing to have time values that are different but `match', so 
first you want to reduce your times to hh:mm  (e.g. with substr).

Then you can cbind() the data frames, and use tapply() to do the 
averaging. E.g.

d1 <- read.table("foo1")
d2 <- read.table("foo2")
d <- cbind(d1, d2)
d$time <- factor(substr(as.character(d$x), 1, 5))
tapply(d$y, d$time, mean)
  14:56  14:62  14:68  15:59  16:05  16:11  16:17  16:23  16:33  16:39
0.3250 0.1110 0.2140 0.2575 0.2545 0.3040 0.2550 0.2630 0.2630 0.3000

There are other ways, e.g.

aggregate(d["y"], list(d$time), mean)

or via ave().


On Tue, 6 Jun 2006, Randy Zelick wrote:

> Hello,
>
> I am trying to average a number of data sets where the x vector contains
> times and the Y data are instrument readings. The vectors all have
> different numbers of values, but many X time values match. For example:
>
> A fragment of the first data set:
>
> x vect    value
> 14:56:10  0.325
> 14:62:11  0.111
> 14:68:11  0.214
> .
> .
> .
> this can go on for 100's of values, spaced by 6 minutes, but there can be
> gaps too, like:
>
> 16:05:18  0.245
> 16:11:09  0.266
> 16:17:05  0.271
> 16:33:00  0.304
> 16:39:05  0.300
>
>
> A fragment of the second data set:
>
> 15:59:08  0.255
> 16:05:44  0.281
> 16:11:25  0.249
> 16:17:39  0.238
> 16:23:51  0.288
>
>
> ...and the result I am looking for is a new vector that looks like this:
>
> 14:56:10  0.325
> 14:62:11  0.111
> 14:68:11  0.214
> .
> .
> .
> 15:59:08  0.255
> 16:05:18  0.2630  * matches, so average
> 16:11:09  0.2575  * matches, so average
> 16:17:05  0.2545  * matches, so average
> 16:33:00  0.304
> 16:39:05  0.300
>
>
> The times values in the new X vector are properly interleaved taking data
> across the datasets, but when there is a match the Y values are averaged.
> Note that the seconds don't matter.
>
> I am using R Version 2.2.1 on a PC
>
> Thanks for thinking about it,
>
> =Randy=
>
> R. Zelick				email: zelickr at pdx.edu
> Department of Biology			voice: 503-725-3086
> Portland State University		fax:   503-725-3888
>
> mailing:
> P.O. Box 751
> Portland, OR 97207
>
> shipping:
> 1719 SW 10th Ave, Room 246
> Portland, OR 97201
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>

-- 
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