Hello all,
I was wondering if there is an R function to do the following:
[*] log(pnorm(x)-pnorm(y)), where x>y.
I don't want all the area under the natural log of the normal pdf less than
x, I only want the area between y and x.
I am aware of the ability to specify log.p=TRUE, which gives me the log of
the probability that X<=x. This does not help me, because the following
code:
pnorm(x, log.p=TRUE)-pnorm(y,log.p=TRUE) is not the same as [*]
mathematically.
I cannot use [*] because some of my x's are far less than the mean, more
than 10 sd. This causes me to take the log(0) which is an error. Thus, I
need to stay in the log scale, since, for z less than 10 sd below the mean,
log(pnorm(z)) is an error, and
pnorm(z,log.p=TRUE) is stable even though theoretically they are equivalent.
Thanks for your time
Justin
[[alternative HTML version deleted]]
______________________________________________
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.
