Forecasting Financial Markets Using High-Frequency Trading Data: Examination with Strongly Typed Genetic Programming

ABSTRACT

Market regulators around the world are still debating whether high-frequency trading (HFT) plays a positive or negative role in market quality. We develop an artificial futures market populated with high-frequency traders (HFTs) and institutional traders using Strongly Typed Genetic Programming (STGP) trading algorithm. We simulate real-life futures trading at the millisecond time frame by applying STGP to E-Mini S&P 500 data stamped at the millisecond interval. A direct forecasting comparison between HFTs and institutional traders indicate the superiority of the former. We observe that the negative implications of high-frequency trading in futures markets can be mitigated by introducing a minimum resting trading period of less than 50 milliseconds. Overall, we contribute to the e-commerce literature by showing that minimum resting trading order period of less than 50 milliseconds could lead to HFTs facing a queuing risk resulting in a less harmful market quality effect. One practical implication of our study is that we demonstrate that market regulators and/or e-commerce practitioners can apply artificial intelligence tools such as STGP to conduct trading behavior-based profiling. This can be used to detect the occurrence of new HFT strategies and examine their impact on the futures market.

KEY WORDS AND PHRASES:

• Evolutionary computation
• artificial intelligence
• high-frequency trading
• algorithmic trading
• big data analytics
• financial econometrics

Supplemental data

Supplemental data for this article can be accessed on the publisher’s website.

Notes

1 Frino et al. [19] use several proxies to identify algorithmic trading in futures markets.

2 Whereas Python and Java programming languages are suitable for trading at the minute time frame, C++, ASIC, and FPGA languages are appropriate for trading at the very low latencies of microseconds and nanoseconds. Machine learning languages such as GP, STGP, and Genetic Algorithms are appropriate for trading signal research and statistical analysis. All of these programming languages are interconnected in HFT.

3 This process is further explained in 'Structure of the Artificial Futures Market and the Differences Between HFTs and Institutional Traders and The Clearing Mechanism and Order Generation for the Artificial Futures Market'.

4 This is calculated as 50% × (118,500/38.50) – 1,000 = 539 contracts.

5 The E-Mini S&P 500 futures market does not involve marker–taker transaction costs, unlike most equity markets.

6 Our policy recommendation is based on trading orders executed at the millisecond interval only. With recent technological improvements in software and hardware, trading orders are executed at the microsecond and even nanosecond intervals. Therefore, our policy recommendation may not be efficient at these time frames.

Additional information

Notes on contributors

Viktor Manahov

VIKTOR MANAHOV ([email protected]; corresponding author) completed his BA (Hons) in Business Studies at the Open University, UK, and then went on to study an MSc in Finance and Investment Management at the University of Aberdeen, UK. He received his PhD titled ‘An investigation of the behaviour of financial markets using agent-based computational models’ from Newcastle University, UK. He is a member of the UK Higher Education Academy and is currently teaching modules related to finance and stock market trading at the University of York, UK. Viktor is also a member of the editorial board of the Review of Behavioral Finance. His research focuses on agent-based modelling and artificial stock markets; genetic programming trading algorithms; stock market forecasts and valuation of securities; high frequency trading techniques; analysis of financial markets behaviour; empirical properties of asset returns; and stylized facts and statistical issues.

Hanxiong Zhang

HANXIONG ZHANG ([email protected]) is a Senior Lecturer in Banking and Finance at Lincoln International Business School, UK. Hanxiong joined the University of Lincoln as a Lecturer in Financial Economics in April 2014. He holds a PhD from University of Newcastle upon Tyne, UK. His research interests cover time-series modelling, market efficiency, trading strategy and asset pricing. He is a CFA charter holder.