In article <VYUQ5.127669$[EMAIL PROTECTED]>,
"Somebody" <[EMAIL PROTECTED]> wrote:
> I wish to study 'streaks' - ie. periods of deviation from expected
results
> based upon large samples.
>
> Could someone please advise as to the proper statistical-mathematical
name
> for 'streaks' and suggest other keywords that would lead me to
discussions
> of this phenomenon?
>
Don't know about keywords, but I'd look through back issues of "Chance"
and "The American Statistician", which publish many articles on
statistics in sports. Maybe these journals/magazines have online
indices, since they're ASA publications. Some time over the past few
years (sorry I can't be more specific), "The American Statistician" had
an article on just this topic. It looked at probabilities associated
with streaks of the kind often mentioned by baseball announcers,
eg "Dave Winfield is hitting .438 over his last 17 games." A number of
papers have studied NBA data to see if the "hot hand" concept is valid.
Sent via Deja.com http://www.deja.com/
Before you buy.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
