A More Accurate Theory of Behavior in Continuous Double Auctions

The continuous double auction (CDA) is a well-studied, widely used institution for trading. But there is no generally accepted theory about the dynamics of price formation within the first period that has survived experimental testing. In this paper, we describe a model of behavior (called IEL-CDA) that builds on the models of Zero Intelligence (Gode and Sunder, 1993) and Individual Evolu- tionary Learning (Arifovic et al., 2022). We use data from our experiments to test the hypothesis that efficiencies and average prices generated by the experiments come from the same distribution as those generated by simulations with the various behavioral models. The hypothesis is rejected for both ZI and IEL; it is not rejected for IEL-CDA. That is, IEL-CDA is a more accurate theory of behavior in the continuous double auction