(1) What has this to do with recursion?
(2) You probably need to use ifelse(). I believe that this is
(in effect) an FAQ.
cheers,
Rolf Turner
On 20/05/11 07:42, Tremblay, Pierre-Olivier wrote:
Hi,
I created a function for obtaining the normal cumulative distribution (I
know all this already exists in R, I just wanted to verify my
understanding of it). below is the code I came up with.
cdf<-function(x) {
erf<-function(x) {
# approximation to the error function (erf) of the
# normal cumulative distribution function
# from Winitzki (2008)
a<- 0.147
partc<- 1+a*x^2
partb<- 4/pi+a*x^2
parta<- -x^2*(partb/partc)
erf<- sqrt(1-exp(parta))
erf
}
# cumulative density function
cdf<- 1/2*(1+erf(x/sqrt(2)))
cdf
}
The erf(x) produces positive values whatever the sign of x. Instead of
obtaining the expected sigmoid shape of the erf() I end up with a
positive values V shape. erf(x) should be negative for negative values
of x. I figure I need some form of conditional statement in the above
code but all my attempts at using "if" and "for" failed miserably.
Clearly, I don't understand how to use them properly in R. If anyone
could point me in the right direction as on how to code conditional
events in R that would be much appreciated.
many thanks in advance
______________________________________________
R-help@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.
