Date: Thu, 20 Mar 2003 17:28:40 -0500
To: WSFA members <WSFAlist@keithlynch.net>
From: Candy Madigan <candymadigan@mindspring.com>
Subject: [WSFA] Re: Predicting who will be at WSFA meetings
Reply-To: WSFA members <WSFAlist@keithlynch.net>

You guys have earned some major geek points with this discussion.

At 12:29 AM 3/20/2003 -0500, you wrote:
>ronkean@juno.com wrote:
>
> > You presumably mean that 5149 is the cumulative recorded attendance
> > over that time period, (but just among those 143 people), much the
> > same way that an airline might report carrying a million passengers
> > in a year, even though many of those million were repeat customers
> > during the year.
>
>Right.  That's not counting people who only showed up once or twice.
>
> > It occurs to me that an instance of non-attendance is also a data
> > point, so there would be 27,456 (143 x 192) data points in all, ...
>
>That's one way to look at it.  But a sparse matrix is best represented
>only by what's present, not by what's absent.  Otherwise, why not list
>all six billion people, most of whom have never been to WSFA, and get
>hundreds of billions of data points?
>
> > Doubtless you will be able to construct some formula which has
> > predictive value, one that has inputs such as past attendance with
> > time weighting, weather and traffic conditions, meeting location,
> > competing events, time of year, time of the month and proximity to
> > major holidays, etc.
>
>Yes, if I went to that much work.  Which I won't.  And such a model
>would be worthless, anyhow.  A model can't contain more complexity
>than the data it attempts to explain, otherwise it can explain
>everything and nothing.  In other words, I could plug in any past
>meeting into such a model, and it would correctly "predict" who would
>be there.  But for future meetings it would be utterly worthless.
>
> > But each attendee is different; ...
>
>It's remotely possible that I will look at men and women separately,
>or Maryland and Virginia residents separately.  But trying to push the
>data any further would be pointless.  What can anyone make of your
>recent attendance, or rather non-attendance:
>
>2000                   |2001                    |2002                    |2003
>J F M A M J J A S O N D|J F M A M J J A S O N D |J F M A M J J A S O N D
>|J F M
>VMMVMVBVMVMMVMVCMVMVPVM|VMVMVMVMVMvMVMMVVMVMVMVM|VMVMVMVMVMVMVMMVVMVCVMVM|VMVxV
>   **         *    **   |      ** * ***    ** * *| * * *  *  *            |
>
>Will you be at the next meeting?  That's up to you.
>--
>Keith F. Lynch - kfl@keithlynch.net - http://keithlynch.net/
>I always welcome replies to my e-mail, postings, and web pages, but
>unsolicited bulk e-mail (spam) is not acceptable.  Please do not send me
>HTML, "rich text," or attachments, as all such email is discarded unread.

Candy