Date of Original Version
Abstract or Table of Contents
In this paper we apply a learning model from machine learning, to a human trading crowd to understand why the no trade theorem was rejected. Our results reveal that trading volume in a continuous double auction market is associated with inverse learning curves. Inverse learning results from adverse selection among market takers and strategic advantageous selection among market makers. In contrast to associating adverse selection with market failure in traditional competitive market theory, the rules of a double auction market efficiently exploit individual differences among the trading crowd to generate both large amounts of trading volume and relatively efficient prices.