C4I Center Seminar Series:
Probabilistic Coherence Weighting
for Improving Expert Forecasts


C4I CENTER SEMINAR SERIES
Dr. Kenneth Olson of the C4I Center presents
“Probablistic Coherence Weighting for
Improving Expert Forecasts”

MIST/C Recording

ABSTRACT

Methods for eliciting and aggregating expert judgment are necessary when decision-relevant data are scarce. This presentation addresses how multiple related individual judgments can be used to improve aggregation of probabilities for a binary event across a set of judges.

The work extends previous efforts that use probabilistic incoherence of an individual’s subjective probability judgments to weight and aggregate the judgments of multiple forecasters for the goal of increasing the accuracy of forecasts. Data from two studies demonstrate good practice for eliciting extra probability judgments that allow adjustment of the judgments and assignment of weights to the judgments. The methods result in improvement of more than 30% over the established benchmark of a simple, equal-weighted averaging of judgments.

BIO

Kenneth C. Olson is a research assistant professor in the Volgenau School of Engineering at George Mason Screen Shot 2013-12-19 at 2.19.29 PMUniversity. He earned his Ph.D. in quantitative psychology from The Ohio State University and completed a fellowship with the Intelligence Community Postdoctoral Research Fellowship Program. He is most knowledgeable about social, cognitive, and motor processes in decision making and has extensive experience educating both the public and professionals on common judgment errors.

His teaching and research provide statistical modeling tools to non-statisticians, including work to develop a structured analytic technique to implement Bayesian networks for intelligence assessments. In 2013, he joined the Center of Excellence in C4I as a core researcher in a federally funded research project called SciCast to develop methods for improving the accuracy of science forecasts.

Date/Time
01/10/2014
1:30 pm - 2:30 pm

Location
Engineering Bldg Room 4705