Artificial Long Memory Effects in Two Agend-Based Asset Pricing Models

This note is concerned with two recent agent-based models of speculative dynamics from the literature, one by Gaunersdorfer and Hommes and the other by He and Li. At short as well as long lags, both of them display an autocorrelation structure in absolute and squared returns that comes remarkably close to that of real data at a daily frequency. The note argues that these long memory effects are to be ascribed to the stochastic specification of the price equation, which given the wide fluctuations in these models unduly fails to normalize the price shocks. Under an appropriate respecification, the long memory completely disappears.