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
Pierre-Olivier
[[alternative HTML version deleted]]
(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
Perhaps this is useful:> x=c(-2,0,2) > sign(x)*abs(x)[1] -2 0 2 ------------------------------------------ Robert W. Baer, Ph.D. Professor of Physiology Kirksville College of Osteopathic Medicine A. T. Still University of Health Sciences 800 W. Jefferson St. Kirksville, MO 63501 660-626-2322 FAX 660-626-2965 ------------------------------------------ Robert W. Baer, Ph.D. Professor of Physiology Kirksville College of Osteopathic Medicine A. T. Still University of Health Sciences 800 W. Jefferson St. Kirksville, MO 63501 660-626-2322 FAX 660-626-2965 -------------------------------------------------- From: "Tremblay, Pierre-Olivier" <potremblay at pharmanet.com> Sent: Thursday, May 19, 2011 2:42 PM To: <r-help at r-project.org> Subject: [R] recursive function> 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 > > Pierre-Olivier > > > [[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. >