New paper studying aggregation mechanisms for crowd predictions

Post date: Nov 11, 2019 1:15:16 PM

The term "wisdom of crowds" is well-recognized at least since the publication of James Surowiecki's book with the same title. It refers to the phenomenon that the aggregated estimates of many different individuals often constitute a surprisingly accurate predictor of the unknown quantity or quality to be estimated. In a recent paper by platform researcher Stefan Palan, co-authored with Jürgen Huber and Larissa Senninger from the University of Innsbruck, the authors study which aggregation mechanism performs best. They compare simple means and medians and more complex, market-based mechanisms. They find that the continuous double auction outperforms all other mechanisms in terms of prediction accuracy.Palan, S., Huber, J., Senninger, L., Forthcoming. Aggregation mechanisms for crowd predictions, Experimental Economics, DOI: 10.1007/s10683-019-09631-0.

Cover of Experimental Economics