The Individual Within the Crowd: How Individual Forecasters Contribute to Ideal Group Forecasts

Explore the intricacies of group forecasting in this 47-minute seminar by Clintin Davis-Stober from the University of Missouri. Delve into the optimal configurations for maximizing group forecast accuracy, considering individual biases and correlated errors. Discover how diversity in individual forecasts linearly trades off with individual accuracy in large forecasting groups. Examine a statistical model that estimates individual forecaster features for aggregate predictions. Understand the conditions for choosing between optimal group weighting and simple averaging of individual forecasts. Learn about the wisdom of the crowd, bias-variance tradeoff, and quantitative reasons behind crowd wisdom. Analyze the system of equations underlying group forecasts and explore simulations demonstrating key concepts. Gain insights into forming wiser crowds and improving collective prediction accuracy.