Harnessing the Wisdom of a Crowd of Probabilistic Forecasters
Overview of the talk:
When probabilistic forecasts are available from multiple forecasters, aggregation can be performed to extract the wisdom of the crowd. This talk will describe aggregation methods that can be applied when there are many forecasters, or a record of past forecast accuracy is not available. The median and trimmed means have been proposed as simple and robust alternatives to the simple average. These methods, as well as a number of new proposals, will be demonstrated with applications to macroeconomics and COVID-19 mortality.
James Taylor (Professor of Decision Science, the Saïd Business School University of Oxford)