R help - Oct 2010 - Where precision change

Hello everyone. 
I need some help to understand when number precision in R is set. For this 
please consider the following example


 for (i in c(2:length(final))){
    sizex <- c(sizex,(final[i]-final[i-1],digits=2))) # round is used to
remove
values that are too small like e-17.
    print(round(final[i]-final[i-1],digits=2))
  }

final[2]-final[1] return something like 4.440892e-16, which means that these two
numbers are the same. They are two but as they were derived from a different 
process they are not the same for precision.

Also the line print(round(final[2]-final[1]),digits=2) returns 0 which is 
correct

When the above loop stops executing inside sizex variable I find the value 
4.440892e-16 which I was not expecting.  As you can see from small code snippet 
before setting the value in the sizex I try to round it. The print gives the 
right value but for some reason it seems that inside the loop the precision in 
sizex is changed.

Can you please help me clarify that?

Best Regards
Alex


      
	[[alternative HTML version deleted]]

Hi

r-help-bounces at r-project.org napsal dne 18.10.2010 11:58:06:
> Hello everyone. 
> I need some help to understand when number precision in R is set. For 
this 
> please consider the following example
> 
> 
>  for (i in c(2:length(final))){
>     sizex <- c(sizex,(final[i]-final[i-1],digits=2))) # round is used to
remove 
> values that are too small like e-17.
>     print(round(final[i]-final[i-1],digits=2))
>   }
> 
> final[2]-final[1] return something like 4.440892e-16, which means that 
these two 
> numbers are the same. They are two but as they were derived from a 
different 
> process they are not the same for precision.
> 
> Also the line print(round(final[2]-final[1]),digits=2) returns 0 which 
is 
> correct
> 
> When the above loop stops executing inside sizex variable I find the 
value 
> 4.440892e-16 which I was not expecting.  As you can see from small code 
snippet 
> before setting the value in the sizex I try to round it. The print gives 
the 

Not having your original data I presume you expect you are rounding when 
adding to sizex but you are not

 c(2,((2^0.5)^2,digits=2))

so either you forgot round or you are not telling us the whole story

Regards
Petr
> right value but for some reason it seems that inside the loop the 
precision in 
> sizex is changed.
> 
> Can you please help me clarify that?
> 
> Best Regards
> Alex
> 
> 
> 
>    [[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.

FAQ 7.31

?all.equal

On Mon, Oct 18, 2010 at 5:58 AM, Alaios <alaios at yahoo.com>
wrote:
> Hello everyone.
> I need some help to understand when number precision in R is set. For this
> please consider the following example
>
>
> ?for (i in c(2:length(final))){
> ? ?sizex <- c(sizex,(final[i]-final[i-1],digits=2))) # round is used to
remove
> values that are too small like e-17.
> ? ?print(round(final[i]-final[i-1],digits=2))
> ?}
>
> final[2]-final[1] return something like 4.440892e-16, which means that
these two
> numbers are the same. They are two but as they were derived from a
different
> process they are not the same for precision.
>
> Also the line print(round(final[2]-final[1]),digits=2) returns 0 which is
> correct
>
> When the above loop stops executing inside sizex variable I find the value
> 4.440892e-16 which I was not expecting. ?As you can see from small code
snippet
> before setting the value in the sizex I try to round it. The print gives
the
> right value but for some reason it seems that inside the loop the precision
in
> sizex is changed.
>
> Can you please help me clarify that?
>
> Best Regards
> Alex
>
>
>
> ? ? ? ?[[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.
>


-- 
Jim Holtman
Cincinnati, OH
+1 513 646 9390

What is the problem that you are trying to solve?