[R] rowSums - am I getting something wrong?

[Ivan Calandra]

Hi Tom,

That's once again the floating point number issue: see FAQ 7.31.
Look at this:
sum(m[161,])
[1] 1
sum(m[161,])==1
[1] FALSE
sum(m[161,])-1
[1] 2.220446e-16

So 0.6+0.3+0.1 is indeed greater than 1

Try this instead:
round(sum(m[161,]))==1
[1] TRUE

HTH,
Ivan


Le 3/7/2011 08:08, Thomas.Salvesen at syngenta.com a écrit :
> I am trying to construct a data set with some sequences for example:
>
> a = seq(0,1,0.1)
>
> m = matrix(nrow = 1331, ncol = 3)
> m[,1] = rep(a,121)
> m[,2] = rep(a,11,each = 11)
> m[,3] = rep(a,1,each = 121)
>
> I realize that there may be better ways of doing this, but this approach demonstrates the problem I'm having.
>
> I then want to get the sum of the rows and delete any row with a sum of greater than 1.  But have a problem with rows containing any combination of the values 0.6, 0.3 and 0.1 as the sum of these is clearly 1, but a request for which rows have a sum greater than 1 will return rows with these values.  Row 161 is the first row containing these values:
>
> [161,]  0.6  0.3  0.1
>
> which(rowSum(m)>1)
>
>> [53]  119  120  121  132  142  143  152  153  154  161  162
> As far as I can tell this only affects combinations of 0.6, 0.3 and 0.1 (though I haven't checked every value in the matrix)
>
> If I try the following:
>
> q=rowSums(m)
> which(q>1)
>
>> [53]  119  120  121  132  142  143  152  153  154  161  162
> But if I add and subtract 1 from this:
>
> q=q+1
> q=q-1
> which(q>1)
>
> [53]  119  120  121  132  142  143  152  153  154  162
>
> What exactly is going on here?  I don't have the problem with other combinations (eg 0.7, 0.2, 0.1).  I assume that there is something about the data format that I don't understand, but if I make a data frame of the matrix I found the same effect.
>
> Any help would be great
>
> Tom
>
>
>
>
>
>
> message may contain confidential information. If you are not the designated recipient, please notify the sender immediately, and delete the original and any copies. Any use of the message by you is prohibited.
>
> 	[[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.
>

-- 
Ivan CALANDRA
PhD Student
University of Hamburg
Biozentrum Grindel und Zoologisches Museum
Abt. Säugetiere
Martin-Luther-King-Platz 3
D-20146 Hamburg, GERMANY
+49(0)40 42838 6231
ivan.calandra at uni-hamburg.de

**********
http://www.for771.uni-bonn.de
http://webapp5.rrz.uni-hamburg.de/mammals/eng/1525_8_1.php