It is a strange behaviour in that I did not expect it... but I am sure there is a simple explanation for it and it'll have to do with the way numbers are stored in R, but it's caught me by surprise and I don't find it obvious.

Here's a simplified example reproducing the behaviour I encountered:

I create an empty vector, and I fill it with a sequence of numbers: 0, 0.005, 0.01, 0.015, 0.02, etc... until 0.1. (In my real code this is just a way to store certain values when certain conditions are met etc etc) Then I ask which element contains teh value 0.01. Ok. Same with 0.02. But when I ask about 0.03 or above... the answer is none. But I can see "0.03" in my vector, it is there!

This must be because of the value stored internally. Indeed, if I do:

v-0.03, when I reach the value that appeared to contain 0.03, I don't obtain zero, but "3.47e-18"

why is this?
what is the recommended way to overcome this?
Interestingly, if I store the values as "character", then I can match them fine...

Below is the code, and my sessionInfo() output.

Thanks!

Jose de las Heras


code example:
#v<-vector(mode="character")
v<-c()
i<-0
while (i <= 0.1)
  {
  v<-c(v,i)
  i<-i+0.005
  }
v
which(v==0.01)
which(v==0.02)
which(v==0.03)



sessionInfo()
R version 2.10.0 (2009-10-26)
i386-pc-mingw32

locale:
[1] LC_COLLATE=English_United Kingdom.1252
[2] LC_CTYPE=English_United Kingdom.1252
[3] LC_MONETARY=English_United Kingdom.1252
[4] LC_NUMERIC=C
[5] LC_TIME=English_United Kingdom.1252

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] biomaRt_2.2.0

loaded via a namespace (and not attached):
[1] RCurl_1.3-1  tools_2.10.0 XML_2.6-0



--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

______________________________________________
R-help@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.

Reply via email to