From: Martin Maechler <maechler_at_stat.math.ethz.ch>

Date: Wed 01 Feb 2006 - 19:42:21 EST

R-help@stat.math.ethz.ch mailing list

https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Wed Feb 01 19:49:56 2006

Date: Wed 01 Feb 2006 - 19:42:21 EST

>>>>> "William" == William Asquith <wasquith@austin.rr.com> >>>>> on Tue, 31 Jan 2006 21:14:50 -0600 writes:

William> I have question (curiosity) regarding returned values of R's qcauchy William> () function, William> for nonexceedance probability (F). It seems the ideal returned range William> of cauchy distribution should be [-Inf,Inf].

William> For F=0

>> qcauchy(0)

William> [1] -Inf

William> but for F=1

>> qcauchy(1)

William> [1] 8.16562e+15

William> It seems to me that the proper return value should be Inf???

yes, but istn't 8 * 10^15 a good approximation to +Inf ? :-) :-)

William> For default (location=0,scale=1) quantile function of cauchy

William> x(F) = tan(pi * (F - 0.5))

William> For F = 0

>> tan(pi*(-0.5))

William> [1] -1.633124e+16

William> For F = 1

>> tan(pi*(0.5))

William> [1] 1.633124e+16

William> So I conclude that qcauchy(0) properly handles the -Inf result and William> the qcauchy(1) returns a very large number, William> curiously not equal to tan (0.5*pi), but certainly not Inf.

The reason for the current behavior is the following part of qcauchy.c :

return location + (lower_tail ? -scale : scale) / tan(M_PI * p); /* -1/tan(pi * p) = -cot(pi * p) = tan(pi * (p - 1/2)) */

(note the comment!)

I'll fix this, since I had wanted to do it in the past (and then thought it wasn't important enough to warrant an extra if() in the code).

Martin Maechler, ETH Zurich

William> As double check,

>> tan(pi*(0.99999-0.5))

William> [1] 31830.99

>> qcauchy(0.99999)

William> [1] 31830.99

R-help@stat.math.ethz.ch mailing list

https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Wed Feb 01 19:49:56 2006

*
This archive was generated by hypermail 2.1.8
: Thu 02 Feb 2006 - 02:13:46 EST
*