From: Spencer Graves <spencer.graves_at_pdf.com>

Date: Sun 03 Jul 2005 - 13:33:20 EST

Date: Sun 03 Jul 2005 - 13:33:20 EST

There are PP plots and QQ plots. I've tried PP plots and didn't get much from them. The normal QQ plot [qqnorm, in R], however, I use for all kinds of things. In a data mining situation, I'll get, e.g., 500 p.values, all uniformily distributed under the null hypothesis. I'll transform that null distribution to N(0, 1) with qnorm(p.values) and make

qqnorm(qnorm(p.values), datax=TRUE).

I may be chastised severely by Tukeyites for "datax=TRUE", but the human eye can decode a 45 degree angle easier than any other angle other than horizontal or vertical. If you have a couple of mild outliers with a typical aspect ratio of the plot, the "datax=TRUE" option will make the line moderately close to 45 degrees.

spencer graves

Jonathan Baron wrote:

> On 07/02/05 15:41, Yulei He wrote:

*> Hi, there.
**>
**> Is there any function in R to plot the probability-probability plot (PP
**> plot)? Suppose I am testing some data against normal.
**>
**> qqnorm might be what you want, or lead to it.
**>
*

-- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA spencer.graves@pdf.com www.pdf.com <http://www.pdf.com> Tel: 408-938-4420 Fax: 408-280-7915 ______________________________________________ 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.htmlReceived on Sun Jul 03 13:36:20 2005

*
This archive was generated by hypermail 2.1.8
: Fri 03 Mar 2006 - 03:33:11 EST
*