Nate Silver is the statistician quoted most often on MSNBC. He provides what appear to be daily updates on what he predicts the electoral college votes will be for each of the two main candidates.
He has a new book out which may interest the list: The Signal and the Noise: Why So many Predictions Fail - But Some Don't. Some nice stuff on climate change in there.
Here's the amazon link:
http://www.amazon.com/Signal-Noise-Many-Predictions-Fail-but/dp/159420411X/ref=sr_1_1?s=books&ie=UTF8&qid=1351647957&sr=1-1&keywords=Nate+Silver
Cheers, Nigel Waters
----- Original Message -----
From: Russell Ecob <[log in to unmask]>
Date: Tuesday, October 30, 2012 5:39 pm
Subject: american elections and statistics of polls in swing states
> Am I alone in being sceptical of the comments in BBC news today
> about the
> key swing state (Ohio) being within the 'margin of error' ? The
> graph (of a
> poll of polls I assume) shows the democratic lead (from memory)
> reducingbut keeping positive with the democratic lead decreasing
> but at a reducing
> rate. I suspect that the statistical test, if that it is, is of
> the poll at
> the most recent instant.
>
> Surely what statisticians should be doing is forecasting the
> opinion poll
> lead, if any, at election day (with appropriate error bands).
> Then, as we
> all know, unpredictable things will happen (e,g, sandy) which
> may well
> change things. Added to this, people may of course vote
> differently to
> what they said they would.
>
> Does anyone know of (American?) statisticians being involved in this
> prediction game?
>
> Russell
>
> ******************************************************
> Please note that if you press the 'Reply' button your
> message will go only to the sender of this message.
> If you want to reply to the whole list, use your mailer's
> 'Reply-to-All' button to send your message automatically
> to [log in to unmask]
> Disclaimer: The messages sent to this list are the views of the
> sender and cannot be assumed to be representative of the range of
> views held by subscribers to the Radical Statistics Group. To find
> out more about Radical Statistics and its aims and activities and
> read current and past issues of our newsletter you are invited to
> visit our web site www.radstats.org.uk.
> *******************************************************
>
******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
Disclaimer: The messages sent to this list are the views of the sender and cannot be assumed to be representative of the range of views held by subscribers to the Radical Statistics Group. To find out more about Radical Statistics and its aims and activities and read current and past issues of our newsletter you are invited to visit our web site www.radstats.org.uk.
*******************************************************
|