January 13, 2009
In current Psychological Science:
The
Wisdom of Many in One Mind:
Improving Individual Judgments With Dialectical Bootstrapping
Stefan M. Herzog and Ralph
Hertwig
Groups are better than individuals at making
judgments about the future and other unknowns, because individual errors tend
to cancel out. Can the power of averaging that underlies this “wisdom of
crowds” be harnessed to improve individual judgments too? This study
shows that averaging an individual’s first estimate with one made later
fosters accuracy. A single mind can thus simulate the wisdom of many.
Leave a Comment » |
Estimation, Forecasting, Groupthink, Wisdom of crowds |
Permalink
Posted by Tim van Gelder
August 19, 2007
Gambling on tomorrow – provocative piece from The Economist on how climate models have to include something like Bayesian priors, and a difficult challenge to which this gives rise.
Lets just hope the climate skeptics don’t seize on this as another pseudo-reason to dismiss climate modelling altogether.
Powered by ScribeFire.
Leave a Comment » |
Bayes, Forecasting, Statistics |
Permalink
Posted by Tim van Gelder
August 18, 2007
Yet more evidence that investment professionals have no competence in their core area of “expertise”.
The Age newspaper does its own survey of experts’ buy and sell recommendations in 2005.
A couple of years later, it turns out you would have done just as well purchasing their “sells” as their “buys”.
The article draws the wrong lesson however. It claims that the moral is to pay as much attention to the sell recommendations as to the buy ones.
The proper lesson is – pay these “experts” no attention at all. And certainly don’t allow them to make any money at your expense.
Powered by ScribeFire.
Leave a Comment » |
Bullshit, Expertise, Forecasting, Investment |
Permalink
Posted by Tim van Gelder
August 15, 2007
“An academic survey of companies appearing on the cover of US business magazines over a twenty-year period says that the magazines usually backed the wrong horse. Comparing share prices of companies 500 days before and 500 days after coverage indicates that “positive stories generally indicate the end of superior performance and negative news generally indicates the end of poor performance”. Ref: The Economist (UK)”
From Brainmail, nowandnext.com
Powered by ScribeFire.
Leave a Comment » |
Forecasting, Media |
Permalink
Posted by Tim van Gelder