Title: News and price returns from threshold behaviour and vice-versa: exact solution of a simple agent-based market model

Abstract: I solve the stationary state of Ghoulmie-Cont-Nadal model of traders with evolving heterogeneous reaction thresholds to news, and obtain exact relationships between news, threshold and price return distributions. For any reasonable news distribution, this model is not able to transform news into fat-tailed price returns. When the variance of the news so small that only players with zero threshold can possibly react to news, this model produces power-law distributed price returns with an exponent of -1. Numerical simulations fully confirm theoretical predictions.