Endogenous cycles in heterogeneous agent models: a state-space approach

This paper proposes an empirical test to identify possible endogenous cycles within heterogeneous agent models (HAMs). We consider a two-type HAM into a standard small-scale dynamic asset pricing framework. Fundamentalists base their expectations on the fundamental value, while chartists consider the level of past prices. Because these strategies, by their nature, cannot be directly observed but can cause the response of the observed data, we construct a state-space model where agents' beliefs are considered the unobserved state components and from which the heterogeneity of fundamentalist-chartist trader cycles can be mathematically derived and empirically tested. The model is estimated using the S&P500 index for the period 1990-2020 at different time scales, specifically, quarterly, monthly, and daily. We find empirical evidence of endogenous damped fluctuations with a higher probability of chartist behavior in the short-term horizon. In addition, the model exhibits better long-run out-of-sample forecasting accuracy compared to the benchmark random walk model.