From fork-admin@xent.com  Tue Aug  6 19:25:57 2002
Return-Path: <fork-admin@xent.com>
Delivered-To: yyyy@localhost.netnoteinc.com
Received: from localhost (localhost [127.0.0.1])
	by phobos.labs.netnoteinc.com (Postfix) with ESMTP id EBE6A440C8
	for <jm@localhost>; Tue,  6 Aug 2002 14:25:56 -0400 (EDT)
Received: from phobos [127.0.0.1]
	by localhost with IMAP (fetchmail-5.9.0)
	for jm@localhost (single-drop); Tue, 06 Aug 2002 19:25:56 +0100 (IST)
Received: from xent.com ([64.161.22.236]) by dogma.slashnull.org
    (8.11.6/8.11.6) with ESMTP id g76IQKk14238 for <jm@jmason.org>;
    Tue, 6 Aug 2002 19:26:23 +0100
Received: from lair.xent.com (localhost [127.0.0.1]) by xent.com (Postfix)
    with ESMTP id 4FEB3294099; Tue,  6 Aug 2002 11:23:04 -0700 (PDT)
Delivered-To: fork@spamassassin.taint.org
Received: from localhost.localdomain (pm3-0.sba1.netlojix.net
    [207.71.218.144]) by xent.com (Postfix) with ESMTP id 3C085294098 for
    <fork@xent.com>; Tue,  6 Aug 2002 11:22:17 -0700 (PDT)
Received: (from dave@localhost) by maltesecat (8.8.7/8.8.7a) id LAA29855;
    Tue, 6 Aug 2002 11:30:11 -0700
Message-Id: <200208061830.LAA29855@maltesecat>
To: fork@spamassassin.taint.org
Subject: S&P QC
From: Dave Long <dl@silcom.com>
Sender: fork-admin@xent.com
Errors-To: fork-admin@xent.com
X-Beenthere: fork@spamassassin.taint.org
X-Mailman-Version: 2.0.11
Precedence: bulk
List-Help: <mailto:fork-request@xent.com?subject=help>
List-Post: <mailto:fork@spamassassin.taint.org>
List-Subscribe: <http://xent.com/mailman/listinfo/fork>, <mailto:fork-request@xent.com?subject=subscribe>
List-Id: Friends of Rohit Khare <fork.xent.com>
List-Unsubscribe: <http://xent.com/mailman/listinfo/fork>,
    <mailto:fork-request@xent.com?subject=unsubscribe>
List-Archive: <http://xent.com/pipermail/fork/>
Date: Tue, 06 Aug 2002 11:30:10 -0700


Begeman's _Manufacturing Processes_
talks about Quality Control Charts:

> The tolerance limits for a part are placed outside of the control
> limits and are never inside of them.  Once the control chart has
> been established, data is recorded on it and it becomes a record
> of the variation of an inspected dimension over a period of time.
> The data plotted should fall in random fashion between the control
> lines, if all assignable causes for variation are absent.  When the
> data fall in this manner, it can be assumed that the part is being
> made correctly 99.73% of the time; that is, no more than 3 out of
> 1000 will leave the process incorrectly made.

Taking a look at the S&P 500, and 
using the Ibbotson 2000 figures of
13% expected stock return and 20%
standard deviation, and taking the
the actual 2.5 year return on the
S&P since 31 Dec 1999 to be close
to -30%, we find:

-.30 ? 2.5*(.13) - 3*sqrt(2.5)*(.20)
-30% > -60%

so it still plots out between the
control lines ...

> So long as the points fall between the control lines, no adjustments
> or changes in the process are necessary.  If five to seven points fall
> on one side of the mean, the process should be checked.  When points
> fall outside of the control lines, the cause must be located and
> corrected immediately.

... but maybe we should've checked
the process during the "long" bull.

-Dave

http://xent.com/mailman/listinfo/fork