How do you guys make your predictions?
We convert betting odds in individual electorates to probabilities of victory. We use these probabilities to simulate elections. From these simulations, we make predictions not just about who will win the 2013 federal election, but precisely where and how they will win. These predictions are difficult to make with standard polling techniques.
What is your comments policy?
This blog is about statistical techniques for predicting election results. We welcome relevant discussion and debate. However, we moderate comments to ensure the debate stays focused and civil. This blog is probably not the best place for intense political discussion. There are many excellent blogs where this is welcome; please use those. Comments that are off-topic or offensive will be removed.
Who are you?
Kaighin McColl is a PhD student in the Department of Civil and Environmental Engineering at MIT. He has degrees in environmental engineering and applied math from the University of Melbourne. A list of Kaighin’s publications can be found here. To follow him on twitter, check out @KaighinMcColl.
Leng Lee runs operations for Codecademy, an internet startup in NYC. He was previously the Director of Analytics at Teach For America and Data Scientist at Codecademy. He has honors degrees in economics and law from the University of Melbourne, and completed a doctorate at Oxford University as a Rhodes Scholar. He has published on the AFL draft, discrimination against migrant workers, and the effect of parental migration on child educational outcomes. Follow him @lenglee_
We’d like to also thank Jeff Chan (MIT) and Bob Chen Ren (UIUC) for their help.