[R] Problem with <= (less than or equal): not giving the expected result

Rainer Johannes Johannes.Rainer at eurac.edu
Tue Apr 5 16:32:31 CEST 2016


On 05 Apr 2016, at 16:07, Thierry Onkelinx <thierry.onkelinx at inbo.be<mailto:thierry.onkelinx at inbo.be>> wrote:

You could use something like this

x <- abs(0.95 - 1)
treshold <- 0.05
x < treshold | abs(x - treshold) < 1e-6

ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
Kliniekstraat 25
1070 Anderlecht

To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of. ~ Sir Ronald Aylmer Fisher
The plural of anecdote is not data. ~ Roger Brinner
The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data. ~ John Tukey

2016-04-05 14:46 GMT+02:00 Rainer Johannes <Johannes.Rainer at eurac.edu<mailto:Johannes.Rainer at eurac.edu>>:
Thanks Adrian and Thierry (from the previous answer).

I was aware of the all.equal function, but there is nothing similar for <= (e.g. all.smallerEqual)?

cheers, jo

On 05 Apr 2016, at 14:31, Adrian Dușa <dusa.adrian at unibuc.ro<mailto:dusa.adrian at unibuc.ro><mailto:dusa.adrian at unibuc.ro<mailto:dusa.adrian at unibuc.ro>>> wrote:

Yes, that does have to do with floating point representation.
I use this function for these types of comparisons (works with values as well as with vectors):

check.equal <- function(x, y) {
    check.vector <- as.logical(unlist(lapply(x, all.equal, y)))
    check.vector[is.na<http://is.na/><http://is.na/>(check.vector)] <- FALSE



On Tue, Apr 5, 2016 at 2:34 PM, Rainer Johannes <Johannes.Rainer at eurac.edu<mailto:Johannes.Rainer at eurac.edu><mailto:Johannes.Rainer at eurac.edu<mailto:Johannes.Rainer at eurac.edu>>> wrote:
Dear All,

I have the following problem:

I have a function in which I check if the difference between values is smaller or equal to a certain threshold. I however realized that I might get there some unexpected results:

> abs(1 - 0.95) >= 0.05
[1] TRUE
## So that’s fine, but:
> abs(1 - 0.95) <= 0.05

Apparently, abs(1 - 0.95) is not equal to 0.05, which I find however quite disturbing.

Along these lines:
> abs(0.95 - 1) > 0.05
[1] TRUE
> abs(0.95 - 1) < 0.05

I guess that has to do with the floating point representation of the data?

Is there something I miss or is there any solution to this?
Thanks for any help!

cheers, jo

I tried this on different R-version (including 3.2.3 and 3.3.0 alpha); The R-version I used for the code above is:

> sessionInfo()
R version 3.0.2 (2013-09-25)
Platform: x86_64-unknown-linux-gnu (64-bit)

 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
 [9] LC_ADDRESS=C               LC_TELEPHONE=C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base
R-help at r-project.org<mailto:R-help at r-project.org><mailto:R-help at r-project.org<mailto:R-help at r-project.org>> mailing list -- To UNSUBSCRIBE and more, see
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html<http://www.r-project.org/posting-guide.html><http://www.r-project.org/posting-guide.html>
and provide commented, minimal, self-contained, reproducible code.

Adrian Dusa
University of Bucharest
Romanian Social Data Archive
Soseaua Panduri nr.90
050663 Bucharest sector 5

        [[alternative HTML version deleted]]

R-help at r-project.org<mailto:R-help at r-project.org> mailing list -- To UNSUBSCRIBE and more, see
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html<http://www.r-project.org/posting-guide.html>
and provide commented, minimal, self-contained, reproducible code.

	[[alternative HTML version deleted]]

More information about the R-help mailing list