SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Politics : View from the Center and Left -- Ignore unavailable to you. Want to Upgrade?


To: Ron who wrote (207755)11/7/2012 1:03:01 PM
From: neolib  Read Replies (1) | Respond to of 540952
 
Interestingly enough, this makes his future job harder. He uses the past performance of individual polls to determine how those same polls should feed into his current model. If the polls that have been missing the most try to learn and get closer to the mean on the next election, he is going to need to try and figure out the impact of that on his own model.

He is better off having a spectrum of polls that disagree and use somewhat different methods as his sources than he is having polls which agree, and use similar methods. What he does need from them is that they stay as consistent in their methods as possible over time. It might seem a bit counter intuitive, but having things out of focus so to speak in sources can allow sharper focus in results, something that is known mathematically for a long time, and is in fact used in some very sensitive and highly accurate equipment (things like optical star trackers and some machine vision systems). Nate is doing the same thing.



To: Ron who wrote (207755)11/7/2012 1:56:45 PM
From: D. K. G.  Read Replies (1) | Respond to of 540952
 
Nate Silver and the Lessons of 2012

ritholtz.com



To: Ron who wrote (207755)11/7/2012 5:53:53 PM
From: koan  Respond to of 540952
 
lol, thanks. Did morning Joe eat any crow?