Tag Archives: models

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

Written by Chris F. Masse on March 9, 2010 — Leave a Comment

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

Posted in Forecasting (Science & Practice), Science | Also tagged data, data crunching, forecasting, forecasting models, forecasts, Nate Silver, polling data, Predictions, statistical models

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

Written by Chris F. Masse on March 9, 2010 — Leave a Comment

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

Posted in All Best Posts Ever, Forecasting (Science & Practice), Science | Also tagged 2008 US presidential elections, Barack Obama, forecasting, forecasting models, Nate Silver, political forecasting, Politics, Psychology, statistical models, statistics, SXSW 2009, US politics, US presidential elections

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

Written by Chris F. Masse on January 22, 2010 — Leave a Comment

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 [...]

Posted in All Best Posts Ever, Analysis (Accuracy & Precision), Analysis (Market Efficiency), Analysis (Meta), Collective Forecasting, Collective Intelligence - Wisdom Of Crowds, Exchanges & Markets, Forecasting (Science & Practice), Market Prices & Probabilities | Also tagged BetFair, betting markets, Collective Forecasting, event derivative markets, forecasting, InTrade, Iowa Electronic Markets, polls, prediction markets, Sports, statistical models, statistics, stats

How Monte Carlo simulations can forecast sales

Written by Chris F. Masse on January 12, 2010 — 1 Comment

“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.“

Posted in Business, Forecasting (Science & Practice) | Also tagged forecasting, Monte Carlo simulations, sales, sales forecasting

Bueno de Mesquita uses mathematics to predict political events.

Written by Chris F. Masse on November 22, 2009 — Leave a Comment

Experts Who Predict the Future
Criticism of Bueno de Mesquita’s Theories

Addendum: Philip Tetlock’s foxes and hedgehogs

Posted in Forecasting (Science & Practice), Politics | Also tagged Bueno de Mesquita, forecasting, forecasts, mathematics, modeling, Politics, predicting, Predictions

Modeling Planet Earth is difficult.

Written by Chris F. Masse on October 18, 2009 — 2 Comments

In defense of Super Freakonomics
A commenter on their blog:
I am a paleoclimatologist, and can only say FINALLY. Someone with economics background understands the difference between reality and a model for reality.
Al Gore, Jim Hansen, Paul Krugman, and many on this page want us to base public policy on models claim to predict the future of [...]

Posted in All Best Posts Ever, Ethics, Forecasting (Science & Practice), History, Research, Science | Also tagged Earth, Earth science, Economics, Freakonomics, Planet Earth, Science, Stephen Dubner, Steven Levitt, Super Freakonomics, SuperFreakonomics

Computer scientists are mainly focused on predictive modelling.

Written by Chris F. Masse on September 7, 2009 — Leave a Comment

“As long as the system can ‘predict’ the outcome on unseen data, things are ok. A black box with perfect predictive performance is great. Explanatory models are rarely the focus.”