Midas Oracle .ORG » forecasting

REVOLUTION: Nate Silver says the Internet changed everything in politics. – [VIDEO]

Chris F. Masse — Tue, 09 Mar 2010 11:23:32 +0000

16:35 into, Nate Silver is asked whether he would open up the data fire hose, one day:

Midas Oracle = About + Archives + Authors + Best + Contact + Information Technology + Links + Probabilistic Predictions + Post-Mortem Predictions + Prediction Exchanges + Prediction Software + Publish Your Post + How To Publish

SXSW: Nate Silver explains how he approached political forecasting for the 2008 US presidential elections. – [VIDEO]

Chris F. Masse — Tue, 09 Mar 2010 10:57:12 +0000

In part #2, he speaks about the books he is writing:

Nate Silver explains how he builds his forecasting models. – [VIDEO]

Chris F. Masse — Tue, 09 Mar 2010 10:02:03 +0000

Via Joe Weisenthal

The relative advantage of prediction markets (over conventional means of forecasting, namely polls and statistical models) is remarkably… SMALL.

Chris F. Masse — Fri, 22 Jan 2010 08:27:38 +0000

Sharad Goel:

In a new study, Daniel Reeves, Duncan Watts, Dave Pennock and I compare the performance of prediction markets to conventional means of forecasting, namely polls and statistical models. Examining thousands of sporting and movie events, we find that the relative advantage of prediction markets is remarkably small. [...]

Given that sports and entertainment markets are among the most mature and successful, our results challenge the view that prediction markets are substantively superior to alternative forecasting mechanisms. Nevertheless, it is certainly possible that there are forecasting applications where either the relative advantage of markets is larger, or that such differences in performance are consequential. Thus, while prediction markets may yet prove to be useful, it would seem the enthusiasm for their predictive prowess has outpaced the evidence.

Prediction Without Markets — by Sharad Goel, Daniel M. Reeves, Duncan J. Watts, David M. Pennock

http://www.cam.cornell.edu/~sharad/papers/pred-wo-markets.pdf

Robin Hanson’s comment:

Whoever said that every prediction market would always be more accurate than any other mechanism? [*] I’d say they are more-accurate more often than they are less-accurate, compared to mechanisms with similar resources. And your plots at the bottom look like they are testing calibration, not accuracy.

Read Daniel Reeves’s answer to Robin Hanson.

[*] Google, doc.

How Monte Carlo simulations can forecast sales

Chris F. Masse — Tue, 12 Jan 2010 17:57:15 +0000

“Monte Carlo simulations run in Excel can transform our limited data sets into statistically valid probability models that give us a much more accurate view into the future.“

One of the key reasons why timing and exact prediction is so difficult is the futility of conventional cause-effect thinking in complex adaptive systems.

Chris F. Masse — Thu, 07 Jan 2010 16:33:40 +0000

Efficient Markets and Pattern Predictions

Via Sean Park

Robin Hanson: Prediction markets are interesting as forums, not methods.

Chris F. Masse — Sun, 03 Jan 2010 10:08:01 +0000

“Prediction markets aren’t about emphasizing ordinary Joes over credentialed bigshots; they are about emphasizing whomever [] tends to be right.”

Forecasting the US economy for 2010 and 2011

Chris F. Masse — Mon, 21 Dec 2009 08:19:42 +0000

Economic forecasting for 2010 and 2011

Time-Series Forecasting

Chris F. Masse — Wed, 16 Dec 2009 17:00:27 +0000

- The different kinds of forecasting software —including time-series forecasting.

- Some forecasting startups make use of the open source and/or SaaS models. I need your feedback on all this. Please, comment below, e-mail me, wave me, or tweet me.

- Lokad – (concurrent time-series forecasting) – Video

- Data Applied – (neural network algorithm)

External Links: Lokad – Data Applied

Good article about corporate forecasting in The Economist (circa February 2009)

Chris F. Masse — Tue, 15 Dec 2009 20:50:18 +0000

The Economist / CFO.com:

The usual trick of plugging figures from operating units into spreadsheets appeals to number-crunchers, but can often generate misleading targets, especially when conditions change fast. [...]

Some companies have formalised this kind of approach by creating “rolling forecasts”. At the end of, say, the first quarter of a financial year, managers forecast the remaining three quarters again and then add an extra quarter’s projections, worrying about only the most important financial variables. Many companies in Europe already use such systems, says Emery Sinclair of Revelwood, an American software-services provider, and firms in America are suddenly taking a lot more interest in them. One benefit of rolling forecasts is that they discourage executives from becoming too fixated on the present at the expense of the future.

Via Paul Hewitt