Explaining Exchange Rate Volatility with a Genetic Algorithm

Motivated by empirical evidence, we construct a model where heterogeneous boundedly rational market participants rely on a mix of technical and fundamental trading rules. The rules are applied according to a weighting scheme. Traders evaluate and update their mix of rules by a genetic algorithm. Already for a low probability of fundamental shocks the interaction between the traders results in a complex behavior of the exchange rates. Simulations of the model produce a high volatility, unit roots of the exchange rate, a fuzzy relationship between news and exchange rate movements, cointegration between the exchange rate and its fundamental, fat tails of returns, a declining kurtosis under time aggregation, and evidence of mean reversion and of cluster in both volatility and trading volume.