Message-ID: <45769B2D.14339.4B6573@localhost>
Date: 2006-12-06T09:27:57Z
From: PIKAL Petr
Subject: summaryBy(): Is it the best option?
In-Reply-To: <682431.28939.qm@web23007.mail.ird.yahoo.com>
Hi
I did not see any answer yet so I try. You can use aggregate, by or
sapply.
Something like
aggregate(soc6a[, your columns], list(hh=hh), sum, na.rm=T)
by(soc6a[, your columns], hh, sum, na.rm=T)
sapply(split(soc6a[, your columns], hh), sum, na.rm=T)
But you have to check speed gain by yourself.
HTH
Petr
On 5 Dec 2006 at 1:30, Werner Wernersen wrote:
Date sent: Tue, 5 Dec 2006 01:30:50 +0100 (CET)
From: Werner Wernersen <pensterfuzzer at yahoo.de>
To: r-help at stat.math.ethz.ch
Subject: [R] summaryBy(): Is it the best option?
> Hi,
>
> since I have quite large tables and the processing
> takes quite a while I am
> curious if I can improve the performance of this
> aggregation somehow: At the
> moment I am using summaryBy from the doBy package
> under R 2.4.0, Win2K.
>
> summaryBy(soc_s6aq5 + soc_s6aq7 + soc_s6aq9 +
> soc_s6aq11 ~ hh +
> comgroup,soc6a,postfix=c("","","",""),FUN=sum,
> na.rm=T)
>
> The data.frame has 124100 rows and 13 cols.
>
> Thanks for any hints!
>
> Werner
>
> ______________________________________________
> 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 and provide commented,
> minimal, self-contained, reproducible code.
Petr Pikal
petr.pikal at precheza.cz