Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: Republican Red

I certainly hope he is right, but I read somewhere that his model also predicted that Bush I would win in 1992.


10 posted on 08/11/2004 2:54:33 PM PDT by QQQQQ
[ Post Reply | Private Reply | To 1 | View Replies ]


To: QQQQQ

Beating the Polls
http://www.hpronline.org/news/2004/05/04/UnitedStates/Beating.The.Polls-672911.shtml

"Fair's predictions have proven remarkably accurate. Using after-the-fact predictions based on real data, Fair's model accurately predicted the winner in all but two elections since 1916. Between 1916 and 1996, the standard error was statistically miniscule.

The worst prediction, George H.W Bush's 1992 re-election effort, was about four percent off, and inaccurately predicted Bush as the winner.

The other time in that span that the model failed was in 1960, when Richard Nixon lost a tight election to John F. Kennedy that was within the Fair model's standard error. In 2000, the Fair model was almost dead-on from a statistical perspective regarding the popular vote, yet failed to predict the winner of the Electoral College.

While Fair's model is remarkably accurate, the times it failed do reveal its limitations. The model acknowledges only two parties because Fair assumes that third party candidates draw about equally from each of the major parties. The Fair model's failure to take third parties into consideration may account for its failure to predict Bush's 1992 loss, when third-party candidate Ross Perot disproportionately hurt Bush, and Al Gore's loss in 2000, when Ralph Nader disproportionately hurt Gore. Also, in testing the model's validity Fair used after-the-fact predictions, which involve known quantities about the economy. In projecting forward, Fair must predict future quantities, which of course are not certain. Finally, like the IEM, Fair's model is national-only and does not predict state-by-state results."


21 posted on 08/11/2004 2:59:35 PM PDT by QQQQQ
[ Post Reply | Private Reply | To 10 | View Replies ]

To: QQQQQ

Was Perot included into the equation?

My wife says 55% for W and over with time left in the evening to watch the Hollywackos commiting suicide, and

She was right with Arnold, 2002 Senate, 2000 Bush, 1992 & 1996 Clinton (arghh), 88 Bush, 80 & 84 Reagan (I'm not making this up)

She predicted Reagan to win in 80 during the 1976 GOP convention!. She's a Southern gal from Kentucky with common sense (lucky me).


29 posted on 08/11/2004 3:05:00 PM PDT by wrathof59 (semper ubi sub ubi)
[ Post Reply | Private Reply | To 10 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson