Day-trading of nikkei 225 index futures based on chaos theory

Abstract

From the perspective that financial market time series display chaotic property, we composed a pilot fund. The amount of this fund is 10 million yen formed by a limited partnership. We applied the local fuzzy reconstruction method based on chaos theory to predict a financial time series; the Nikkei 225 index futures market price. And we actually traded those index futures daily to produce a track record during the six months from 1 April 2002 to 30 September 2002. This paper reports the prediction, trading method, trading results, salient problems; expected annual return is 12.0% but actual return is -17.6% including brokerage commission, and discusses its cause and countermeasure.