Statistical arbitrage trading strategy in commodity futures market with the use of nanoseconds historical data

Abstract

This paper confirms the existence of statistical arbitrage opportunities by employing the nanosecond historical data in high frequency trading (HFT). When considering the possible options, the Daniel Herlemont pairs trading strategy has been selected. In order pairs trading could operate, the pair selection algorithm had to be developed. Herlemont pairs trading strategy has not been tested before by using the nanosecond information and the proposed pair selection algorithm. The main objective of the given research is to test the pairs trading strategy in HFT by calculating the returnability in commodity futures market. The statistical arbitrage strategy attempts to achieve profit by exploiting price differences of the futures contracts. The strategy takes long/short positions when the spread between the prices widens with an expectation that the prices will converge in the future. In the given paper, the nanosecond historical data was provided by the Nanotick Company. The applied strategy has been subsequently tested with MatLab software.