s-news
[Top] [All Lists]

Re: Quartiles

To: ripley@stats.ox.ac.uk (Prof Brian Ripley)
Subject: Re: Quartiles
From: "Edward Malthouse" <ecm@casbah.it.northwestern.edu>
Date: Fri, 7 Dec 2001 14:03:04 -0600 (CST)
Cc: s-news@lists.biostat.wustl.edu
In-reply-to: <Pine.LNX.4.31.0112071905480.1544-100000@gannet.stats> from "Prof Brian Ripley" at Dec 07, 2001 07:09:18 PM
I neglected to tell what desirable property 5 in Hyndman and Fan was
in my previous email (the Splus definition of quantiles does not
satisfy this one but the one that SPSS seems to use and PCTLDEF=4 in
SAS do satisfy it):

Where \hat{Q}_i^{-1}(x) (inverse of quantile definition i) is uniquely defined, 
\hat{Q}_i^{-1}(X_{(1)}) > 0 and \hat{Q}_i^{-1}(X_{(n)}) < 1

"Property P5 reflects the result that for a continuous distribution,
we expect there to be positive probability for values beyond the
range of the data."  (p. 362)

Ed Malthouse


Dr. Edward C. Malthouse
Assistant Professor
Integrated Marketing Communications Department
Medill School of Journalism
1908 Sheridan Road
Evanston, IL  60208-1290
Tele:  847-467-3376
Fax:  847-491-5925

<Prev in Thread] Current Thread [Next in Thread>