Discovering trading rules with genetic algorithms: an empirical study based on GARCH time series

摘要

In this paper, the performance of canonical GA-based trading strategies are euvaluated under a well-known nonlinear time seris in finance, namely, the GARCH process. Unlike many existing applications of computational intelligence in financial engineering, for each performance criterion, we provide a rigorous asymptotic statistical test based on Monte Carol simulation. As a re-sult, this study provides us with a thorough under-standing about the effectiveness ofcanonical GAs for evolving trading strategies under the GARCH financial time series.