Skip to content

Looping syntax?

3 messages · Michaell Taylor, Thomas Lumley, Claudia Tebaldi

Original
Michaell Taylor
# 
Sorry for what I hope is an unchallenging question for R veterans.

I have a question regarding looping syntax, though I know this should be 
avoided if possible in R - as in most other stat languages with which I am 
familiar.  

I would like to perform a series of operations on the subsets as defined by 
condition. Given the following data frame 

condition	X 	Y
aa		4	5
aa		8	9
.
.
bb		8	4
bb		3 	9
. 
. 
more ..

There are over 100 condition states, each with several hundred observations.  
For ease of this example, lets assume that I wish to make XY plots of each of 
the 100 condition states, with In other words, I need a structure like the 
following

for state in condition:
	identifier <- c("figure N:", cond)
	postscript(identifier,height=3,width=6,horizontal=F) 
	plot(X,Y, main=indentifier)
	n <- n+1

(Please excuse the bastardized R syntax, but not knowing the R syntax is the 
reason I am now writing)  The goal of this structure is to produce a graph 
for each unique value of condition which are sequentially numbered and 
entitled and saved with the condition state as part of the filename/title of 
the plot.

I know this must be simple in R, but the syntax is throwing me.  As an aside, 
could I get recommendations on the best S-Plus/ R book for data manipulation. 
 I have Venables & Ripley, "Modern Applied Statistics with S-Plus" and a 
smattering of web docs, but nothing seems to go into the details of indexing, 
built in fuctions, etc that seem pretty important.

Thanks in Advance.

Michaell Taylor

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Thomas Lumley
#
On Fri, 29 Jun 2001, Michaell Taylor wrote:

            

      
      
      
    

        
I think you want something like

for(state in unique(condition)){
	postscript(paste("figure",state,sep=""),height=3,horiz=FALSE)
	these<-condition %in% state
	plot(X[these],Y[these],xlab="X",ylab="Y",
		main=paste("Condition is",state))
	dev.off()
}


The loop could be replaced with sapply(), but I would expect that the file
access would be the rate-limiting step so sapply() wouldn't help a lot.
The whole thing could also be done with tapply() or by(). This is left as
an exercise for the reader.


	-thomas

Thomas Lumley			Asst. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Claudia Tebaldi
    

    

      
      #
    

  


  

      
On Fri, 29 Jun 2001, Michaell Taylor wrote:

        

            

      
      
      
    

        
say your dataframe is called pippo and its columns are called 
condition, X, Y by

names(pippo)_c("condition","X",Y")


then 


 unique.cond_unique(pippo$condition) 
 l_length(unique.cond)  

 for(i in 1:l){
 filename_paste("figure_",unique.cond[i],".ps",sep="")  #will be 
                                                        #something
                                                       #like figure_aa.ps
 fig.title_paste("figure N:",unique.cond[i])
 postscript(file=filename,height=3, width=6, hor=T)
 plot(cbind(pippo$X,pippo$Y)[pippo$condition==unique.cond[i],],
	main=fig.title,xlab="X", ylab="Y") 
 dev.off()

}

this gives you as many files as there are unique conditions.


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._