[R] questions about R

Torsten Hothorn Torsten.Hothorn at rzmail.uni-erlangen.de
Thu Jul 15 08:41:55 CEST 2004


On Thu, 15 Jul 2004, [big5] house-ball wrote:

> Hi,
>
> I try to compute critical value for multivariate normal distribution, and I find the crit(fit, const = c(0, 1), d = 1, cov = 0.95, rdf = 0) which seems to compute critical value. However, I can't compute right critical value for multivariate normal distribution which I want. I don't know how to solve my question. I send my question in the file. Would you help me to solve it, please? My English is not very well, but I appreciate you very much.
>

one possibility is to invert the P-value function `pmvnorm' from
package `mvtnorm' using `uniroot' - There is FORTRAN 90 code solving this
problem available from Alan Genz' homepage at

http://www.math.wsu.edu/math/faculty/genz/homepage

Best,

Torsten

>                                                                                Chia-Yi



More information about the R-help mailing list