[R] [Fwd: modes again]
Gregor GORJANC
gregor.gorjanc at bfro.uni-lj.si
Fri Nov 26 10:45:39 CET 2004
Hello!
Ross thanks for the code. I have made a comparison with SAS, and your
code (simpleMode, which is also pasted bellow) works well for
categorical as well as for continous data. However there are problems
with multiple modes.
I have sent this mail also to the R-help list so everybody can benefit.
I hope you do not mind.
---------------------------------------------------------------
#returns the lowest mode of a value set
#finds the mode of a data set using the sort function
#call with source simplemode(var)
simpleMode=function(var)
{
#create a table with the values sorted by frequency
sortvar=sort(table(var))
#put the most freq value (the last one) and its count into a variable
modeTable=sortvar[length(sortvar)]
#get the name (value) of the most frequent value
themoderesult=names(modeTable)
#get the count of the most frequent value (one could use min or max, but
since there is only one, this method is simpler)
maxfreq=as.vector(modeTable)
cat("Mode: ",themoderesult, "(frequency: ",maxfreq,")")
}
-------- Original Message --------
Subject: modes again
Date: Mon, 22 Nov 2004 08:49:54 -0800
From: Ross Nelson <rnelson at cariboo.bc.ca>
To: gregor.gorjanc at bfro.uni-lj.si
Attached are two short functions to generate modes.
Place them in your working directory.
call the simpleMode file with
source("simpleMode")
simpleMode(var)
where var is a vector variable.
call the mode2 file with
source("mode2")
mode2(var)
where var is a vector variable.
A description of the mode2 function is available from
http://cran.r-project.org/doc/contrib/Lemon-kickstart/kr_dstat.html
I wrote the other.
--
Lep pozdrav / With regards / Con respeto,
Gregor GORJANC
---------------------------------------------------------------
University of Ljubljana
Biotechnical Faculty URI: http://www.bfro.uni-lj.si
Zootechnical Department mail: gregor.gorjanc <at> bfro.uni-lj.si
Groblje 3 tel: +386 (0)1 72 17 861
SI-1230 Domzale fax: +386 (0)1 72 41 005
Slovenia
More information about the R-help
mailing list