Quantitative Research in High Frequency Trading for Natural Gas Futures Market

摘要

High frequency trading (HFT) in micro or milliseconds has recently drawn attention of financial researches and engineers. In nowadays algorithmic trading and HFT account for a dominant part of overall trading volume. The main objective of this research is to test statistical arbitrage strategy in HFT natural gas futures market. The arbitrage strategy attempts to profit by exploiting price differences between successive futures contracts of the same underlying asset. It takes long/short positions when the spread between the contracts widens; hoping that the prices will converge back in the near future. In this study high frequency bid/ask and last trade records were collected from NYMEX exchange. The strategy was back tested applying MatLab software of technical computing. Statistical arbitrage and HFT has given positive results and refuted the efficient market hypothesis. The strategy can be interesting to financial engineers, market micro-structure developers or market participants implementing high frequency trading strategies.