[R] Odp: use of class variable in r as in Proc means of sas

[Petr PIKAL]

Hi


r-help-bounces at r-project.org napsal dne 22.09.2009 11:51:18:

> 
> Hi,everyone i need to calculate quartile values of a variable grouped by 
the
> other variable .
> same as in aggregate function(only median,mean or functions is 
possible-i
> think so)
> Could you please help me to achieve the  same for other quartile
> values(5,10,25,75,90) as for median using aggregate.
> Thanks in advance.
> data :
> zip             price
> 60000   567000 
> 60001   478654
> 60004      485647
> 60001      2783958
> 60005      97845848
> 60006      378383478
> 60002      397895735
> 60001      487587575
> 60002      478848
> 60003      49847874
> 60004      467648
> 60005       567489
> 60006       4776746
> 60004       4843949
> what i want i quartile values of price grouped by zip

If you need several values of quantiles together aggregate is not suitable 
as the function needs to return scalar

aggregate(rnorm(60), list(sample(1:5,60, replace=T)), function(x) 
quantile(x,c(.05,.5,.95)))
Error in aggregate.data.frame(as.data.frame(x), ...) : 
  'FUN' must always return a scalar

but you can use by

by(rnorm(60), list(sample(1:5,60, replace=T)), function(x) 
quantile(x,c(.05,.5,.95)))

or some other aggregating functions e.g. in doBy or plyr packages or to 
use split/sapply construct.

Regards
Petr



> -- 
> View this message in context: 
http://www.nabble.com/use-of-class-variable-in-
> r-as-in-Proc-means-of-sas-tp25530654p25530654.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> 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.