出版社：Vilnius University, University of Latvia, Latvia University of Agriculture, Institute of Mathematics and Informatics of University of Latvia
摘要：Nowadays, it is not necessary for humans to conduct trade; this task is performed by
trading algorithms. The speed of trading is of the most importance, however, there are relatively
few aoutademic researches on the increased speed of trading from milliseconds to nanoseconds. In
order to address the aforementioned shortcoming, this research measures the differences in the
effectiveness of the pairs trading strategies, emerging when microsecond and nanosecond data are
included. The effect of the increased speed of data is analysed. We present different pairs trading
strategies and one pair selection algorithm, based on the cointegration method. These trading
strategies are implemented on five different commodity futures contracts using both microsecond
and nanosecond historical data. The effectiveness is measured in accordance with the profit,
generated at the end of the trading period. In order to measure the effectiveness of all presented
pairs trading strategies, the Sharpe Ratio method was introduced.
