UnThreaded | Threaded | Whole Thread (6) | Ignore Thread Prev Thread | Prev | Next | Next Thread
Author: tamhas Big gold star, 5000 posts Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: of 73047  
Subject: Re: More Polls...... Date: 10/8/2012 5:37 PM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 2
I mention Mr. Silver because he is *way* ahead of every one else that is compiling poll data in the sophistication of his analysis and of the model. You should read his book
http://www.amazon.com/Signal-Noise-Many-Predictions-Fail-but...
to learn about the science of prediction and why it often doesn't do very well.

There are two ironic things about your link.

One is that right there on the right side of the page is an electoral college forecast based on poll data which shows a comfortable Obama win, if not quite so dominant as it was pre-debate.

The second is that this model in the link is purely an economic model which does not consider poll data. I.e., it is exactly the kind of model which has very poor predictive value for a couple of reasons. One, all it is considering is economic data and that is not the only thing an American voter considers. Second, it is a classic piece of data mining, i.e., through a bunch of variables into the hopper with no particular theory, but over a relatively limited data set, and see which variables predict the known results. Then, apply that weighting to the new case to make a prediction. This is a problem because the limited data set means that it is highly sensitive to exactly which data one includes. Add a couple more elections and the weighting is quite likely to come out quite differently. Thus, such models have a very poor predictive record. Third, if one looks back at the history of people proposing various kinds of models based on fundamentals, i.e., not polling data, it is abundantly clear that, at best, they do much poorer than actual polling data. They are fun to look at a year out or whatever when there is little polling data because actual candidates are not yet identified, but once one has actual polling data it is far more predictive.

You were clearly trying to depress Mitt voters a few weeks ago

What BS. I may have been trying to pop a couple of hot air balloons, but I really have no need to try to depress voters of any persuasion. As I have said before, if my interest was trying to influence voters, I can think of a long list of things to do that would have more impact than arguing with you lot. I have no illusion that I am going to change anyone's vote by anything I say here.
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Print the post  
UnThreaded | Threaded | Whole Thread (6) | Ignore Thread Prev Thread | Prev | Next | Next Thread

Announcements

Pencils of Promise - Back to School Drive
"Pencils of Promise works with communities across the globe to build schools and create programs that provide education opportunities for children."
Post of the Day:
Value Hounds

Netflix Riles Investors
What was Your Dumbest Investment?
Share it with us -- and learn from others' stories of flubs.
When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Community Home
Speak Your Mind, Start Your Blog, Rate Your Stocks

Community Team Fools - who are those TMF's?
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and "#1 Media Company to Work For" (BusinessInsider 2011)! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.
Advertisement