This paper extends the genetic programming techniques developed in Neely, Weller and Dittmar (1996) to show that technical trading rules can make use of information about intervention by the Federal Reserve to improve their out-of-sample profitability. A considerable part of the improvement in performance results from more efficient use of the information in the past exchange rate series. We show that much of the profitability of the rules is accounted for by returns from t - 1 to t, when intervention takes place at date t. This supports the view that intervention is intended to check or reverse strong and predictable trends in the market. However, the rules interpret intervention as a signal that an existing trend will continue, rather than undergo a reversal.