Prediction Markets + Market Predictions = Collective Forecasting That Pays Off
Yet another reason why we should *not* call it the “Prediction Market Institute”…
So what about a system to evaluate us prognosticators? Professor Tetlock suggests that various foundations might try to create a “trans-ideological Consumer Reports for punditry,” monitoring and evaluating the records of various experts and pundits as a public service. I agree: Hold us accountable!
Should we set up a registry of track records —in addition to our prediction market stuff?…
You can follow all the replies to this entry through the comments feed.
Jason Ruspini on Nate Silver: InTrade probability on ObamaCare is quite right.Jason Ruspini on “In 1965, Congress said Medicare would cost $9 billion by 1990. In reality, it cost $67 billion –seven times more than the prediction.”
Paul Hewitt on The Real Arithmetic Of The Health Care Reform
Chris F. Masse on “In 1965, Congress said Medicare would cost $9 billion by 1990. In reality, it cost $67 billion –seven times more than the prediction.”Chris F. Masse on “In 1965, Congress said Medicare would cost $9 billion by 1990. In reality, it cost $67 billion –seven times more than the prediction.”Chris F. Masse on “In 1965, Congress said Medicare would cost $9 billion by 1990. In reality, it cost $67 billion –seven times more than the prediction.”Paul Hewitt on “In 1965, Congress said Medicare would cost $9 billion by 1990. In reality, it cost $67 billion –seven times more than the prediction.”Chris F. Masse on “In 1965, Congress said Medicare would cost $9 billion by 1990. In reality, it cost $67 billion –seven times more than the prediction.”Paul Hewitt on “In 1965, Congress said Medicare would cost $9 billion by 1990. In reality, it cost $67 billion –seven times more than the prediction.”
Max Keiser on Nate Silver: InTrade probability on ObamaCare is quite right.
Who needs pundits’ track records when we have prediction markets?
http://www.midasoracle.org/2009/03/26/pundits-track-records-prediction-markets/