[R] Modular Arithmetic Error?

iuhz7j202 at sneakemail.com iuhz7j202 at sneakemail.com
Fri Apr 17 16:15:16 CEST 2009


Hi,

I'm using the '%%' operator in some code, and am running into the following erroneous outcome:

  > 1.2 %% 0.2
  [1] 0.2

Unless I'm very mistaken, the result should be 0 (indeed, 12 %% 2 does result in 0). Furthermore:

  > 1.20000000000000001 %% 0.2
  [1] 0.2
  > (1.2+1e17) %% .2
  [1] 0
  Warning message:
  probable complete loss of accuracy in modulus 

(Warning comes up only when it is giving a closer-to-correct answer)
What makes it even stranger is this result:

  > (1.2 %% 0.2) == 0.2
  [1] FALSE

Does anybody know what's going on here? This seems like very bad behavior.

Thanks,
Peter




More information about the R-help mailing list