Re: [R] ?to calculate sth for groups defined between points in one variable (string), / value separating/ spliting variable into groups by i.e. between start, NA, NA, stop1, start2, NA, stop2

From: Joris Meys <jorismeys_at_gmail.com>
Date: Thu, 24 Jun 2010 15:14:21 +0200

On Thu, Jun 24, 2010 at 1:18 PM, Eugeniusz Kaluza <Eugeniusz.Kaluza_at_polsl.pl> wrote:
>
> Dear useRs,
>
> Thanks for any advices
>
> # I do not know where are the examples how to mark groups
> #  based on signal occurence in the additional variable: cf. variable c2,
> # How to calculate different calculations for groups defined by (split by occurence of c2 characteristic data)
>
>
> #First example of simple data
> #mexample   1      2    3  4     5  6  7  8  9  10 11       12 13 14 15 16 17
> c0<-rbind( 1,      2 , 3, 4,      5, 6, 7, 8, 9,10,11,      12,13,14,15,16,17     )
> c0
> c1<-rbind(10,     20 ,30,40,     50,10,60,20,30,40,50,      30,10, 0,NA,20,10.3444)
> c1
> c2<-rbind(NA,"Start1",NA,NA,"Stop1",NA,NA,NA,NA,NA,NA,"Start2",NA,NA,NA,NA,"Stop2")
> c2
> C.df<-data.frame(cbind(c0,c1,c2))
> colnames(C.df)<-c("c0","c1","c2")
> C.df
>
> # preparation of form for explaining further needed result (next 3 lines are not needed indeed, they are only  to explain how to obtain final result
>  c3<-rbind(NA,"Start1","Start1","Start1","Start1","Start2","Start2","Start2","Start2","Start2","Start2","Start2","Start2","Start2","Start2","Start2","Start2")
>  c4<-rbind(NA, "Stop1", "Stop1", "Stop1", "Stop1", "Stop2", "Stop2", "Stop2", "Stop2", "Stop2", "Stop2", "Stop2", "Stop2", "Stop2", "Stop2", "Stop2", "Stop2")
>  C.df<-data.frame(cbind(c0,c1,c2,c3,c4))
>  colnames(C.df)<-c("c0","c1","c2","c3","c4")
>  C.df$c5<-paste(C.df$c3,C.df$c4,sep="-")
>  C.df
>
Now this is something I don't get. The list "Start2-Stop2" starts way before Start2, actually at Stop1. Sure that's what you want?

I took the liberty of showing how to get the data between start and stop for every entry, and how to apply functions to it. If you don't get the code, look at

?lapply
?apply
?grep

I also adjusted your example, as you caused all variables to be factors by using the cbind in the data.frame function. Never do this unless you're really sure you have to. But I can't think of a case where that would be beneficial...

...
C.df<-data.frame(c0,c1,c2)
C.df

# find positions
Start <- grep("Start",C.df$c2)
Stop <- grep("Stop",C.df$c2)

# create indices
idx <- apply(cbind(Start,Stop),1,function(i) i[1]:i[2]) names(idx) <- paste("Start",1:length(Start),"-Stop",1:length(Start),sep="")

# Apply the function summary and get a list back named by the interval. out <- lapply(idx,function(i) summary(C.df[i,1:2])) out

If you really need to start Start2 right after Stop1, you can use a similar approach.

Cheers
Joris

> # NEEDED RESULTS
>  # needed result
> # for Stat1-Stop1: mean(20,30,40,50)
> # for Stat2-Stop2: mean(c(10,60,20,30,40,50,30,10,0,NA,20,10.3444), na.rm=T)
> #mean:
>         c1     c3    c4           c5
>         20  Start1 Stop1 Start1-Stop1
>   25.48585  Start2 Stop2 Start2-Stop2
>
> #sum
> # for Stat1-Stop1: sum(20,30,40,50)
> # for Stat2-Stop2: sum(c(10,60,20,30,40,50,30,10,0,NA,20,10.3444), na.rm=T)
> #sum:
>         c1     c3    c4           c5
>        140  Start1 Stop1 Start1-Stop1
>   280.3444  Start2 Stop2 Start2-Stop2
>
> # for Stat1-Stop1: max(20,30,40,50)
> # for Stat2-Stop2: max(c(10,60,20,30,40,50,30,10,0,NA,20,10.3444), na.rm=T)
> #max:
>         c1     c3    c4           c5
>        50  Start1 Stop1 Start1-Stop1
>        60  Start2 Stop2 Start2-Stop2
>
> # place of max  (in Start1-Stop1: 4 th element in gruop Start1-Stop1
> # place of max  (in Start1-Stop1: 2 nd element in gruop Start1-Stop1
>
>        c0     c3    c4           c5
>         4  Start1 Stop1 Start1-Stop1
>         2  Start2 Stop2 Start2-Stop2
>
>
> Thanks for any suggestion,
> Kaluza
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> 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.
>

-- 
Joris Meys
Statistical consultant

Ghent University
Faculty of Bioscience Engineering
Department of Applied mathematics, biometrics and process control

tel : +32 9 264 59 87
Joris.Meys_at_Ugent.be
-------------------------------
Disclaimer : http://helpdesk.ugent.be/e-maildisclaimer.php

______________________________________________
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.
Received on Thu 24 Jun 2010 - 13:29:24 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Thu 24 Jun 2010 - 15:20:35 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive