Hi All,
I did a google search and could not find the answer. Thus I post this
message. I found runmax only work for positive numbers.
x = rep(-1,10)
runmax(x,3)
x = rep(0, 10)
runmax(x,3)
for 32-bit R, i got very small numbers: 2.121996e-314
for 64-bit R, i got NaN.
Is it a bug in runmax? or should this be included in the documentation that
runmax only for positive numbers?
however, we do have a workaround -- runmin works, thus we can do y
-runmin(-x,3).
Thanks.
jiucang
--
View this message in context:
http://www.nabble.com/runmax-function-only-for-positive-numbers--tp24926641p24926641.html
Sent from the R help mailing list archive at Nabble.com.
Jiucang Hao wrote:> Hi All, > > I did a google search and could not find the answer. Thus I post this > message. I found runmax only work for positive numbers. > x = rep(-1,10) > runmax(x,3) > > x = rep(0, 10) > runmax(x,3) > > for 32-bit R, i got very small numbers: 2.121996e-314 > for 64-bit R, i got NaN. > > Is it a bug in runmax? or should this be included in the documentation that > runmax only for positive numbers? > > however, we do have a workaround -- runmin works, thus we can do y > -runmin(-x,3). >You are talking about runmax in the caTools package? You should let the maintainer know. I've cc'd him in this message. Duncan Murdoch