High-frequency Trading Strategies Research Articles

A comparative study of ensemble learning algorithms for high-frequency trading

High-Frequency Trading utilizes powerful mathematical algorithms to execute transactions at an extremely rapid pace, which makes the use of machine learning techniques for prediction necessary. This paper evaluates the effectiveness of various ensemble learning algorithms, including Boosting (Adaboost and XGBoost), Bagging (Random Forest and Bagging-LSVM), and Stacking, in predicting stock prices using High-Frequency Trading (HFT) data from the Casablanca Stock Exchange. The data contains 311.812 transactions at the millisecond precision level and was obtained under a non-disclosure agreement. Three performance metrics are used: Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Mean of the Squared Errors (MSE) and accuracy of predictions is assessed daily, monthly and annually. The study shows that the Stacking Model outperforms other algorithms in forecasting prices across different periods due to its ability to generate profit from multiple learners and construct a robust model. These results provide valuable insights for improving High-Frequency Trading (HFT) strategies, including the accuracy and robustness of the algorithms in predicting stock prices, which are crucial for effective HFT strategies. The findings highlight how each algorithm characteristics and performance metrics can enhance decision-making in HFT. The novelty of this study lies in its unique combination of a focus on an emerging market, the application of advanced ensemble learning algorithms to high-frequency data, and its comprehensive approach to evaluating algorithm performance. Also, this paper is the first study that focuses on applying ensemble learning algorithms to high-frequency data in Casablanca Stock Exchange and one of the rare that focuses on African stock markets. Thus, this study is pioneering in its comparative analysis of machine learning techniques for HFT in Morocco, introducing sophisticated and more accurate tools for analyzing and predicting Casablanca Stock Exchange trends.

Market price determination: Interpreting quote order imbalance under zero-profit equilibrium

We address the challenge of accurately determining market prices in order-driven markets, where the granularity of price adjustments is set by exchange rules, such as minimum increments of 0.01. Our approach advances traditional pricing models in accuracy by interpreting the dynamics of quote order imbalance, defined as the difference in volume between buy and sell orders at the best available prices, within a zero-profit equilibrium framework. This framework assumes that market competition leads to a state where no additional profit can be made from these imbalances. Analyzing high-frequency data from China's A-share market, we demonstrate the variable impact of order imbalances on price formation, influenced by the depth and liquidity of the market. We also uncover a potential threshold effect of order flow autocorrelation on pricing. Our findings enhance market microstructure research and high-frequency trading strategies by offering a valuable tool for understanding price formation's complex dynamics.

The Flash Crash: The Impact of High-Frequency Trading on the Stability of Financial Market

The growing prevalence of High-Frequency Trading (HFT) in financial markets has sparked intense debate over its impact on market stability. This paper aims to research the adverse effects that come along with HFT's enhancement of liquidity and price discovery efficiency. It contrasts the hardware infrastructure and execution speeds of high-frequency automated trading against manual trading and analyzes how the liquidity bubbles provided by HFT contribute to market volatility. This paper then concludes with three main points. First, HFT can pose a threat to market stability. Second, specific HFT strategies may mislead other market participants and lead to price distortions. Lastly, HFT may exacerbate market participation inequality, posing a market fairness challenge. Based on these findings, this paper recommends that regulatory authorities and policymakers pay closer attention to the potential risks of HFT and implement measures to maintain market stability and fairness.

A Novel Window Analysis and Its Application to Evaluating High-Frequency Trading Strategies

A Novel Window Analysis and Its Application to Evaluating High-Frequency Trading Strategies

High-Frequency Trading Using Machine Learning: A Comprehensive Analysis

This research paper delves into the intersection of High-Frequency Trading (HFT) and Machine Learning (ML), exploring the significant impact of ML techniques on enhancing the efficiency, accuracy, and profitability of HFT strategies. The paper presents an in-depth examination of the principles, challenges, and opportunities associated with HFT and ML integration. It also discusses various ML approaches applied in HFT, their advantages, limitations, and potential future developments. Through an extensive review of literature and case studies, this paper aims to provide a comprehensive overview of the evolving landscape of HFT driven by ML advancements.

Fractal Geometry in High-Frequency Trading: Modeling Market Microstructure and Price Dynamics

This theoretical article delves into the intricate world of high-frequency trading (HFT) without empirical testing of real-world data, focusing on the incorporation of fractal geometry principles to enhance our understanding of market microstructure and price dynamics. In the introduction, we outline the significance of this research in the context of modern financial markets and lay out the objectives of our theoretical analysis. The article then takes an in-depth dive into fractal geometry fundamentals, illuminating its core concepts and its relevance within financial markets. Subsequently, the article explores the landscape of high-frequency trading, offering an overview of this dynamic domain and how fractal geometry can be incorporated into trading models. The section on modeling market microstructure presents theoretical approaches to understanding order flow dynamics, including novel derivations and equations. It then transitions into fractal-based approaches for analyzing the complexities of market microstructure, providing both an original perspective and numbered equations. Moreover, this article investigates the theoretical modeling of price dynamics, underscoring the pivotal role of fractal geometry in enriching these models. The discussion revolves around the fundamental autoregressive models and multifractal models, and it elucidates how fractal geometry principles, such as the Hurst exponent, come into play. We explore the self- similarity of price dynamics, fractal dimensions, and how these aspects can be integrated into high-frequency trading strategies. Overall, this article offers a comprehensive theoretical exploration of fractal geometry's implications in the realm of high-frequency trading, providing valuable insights for both researchers and practitioners seeking to fathom the complexities of market microstructure and price dynamics. The incorporation of fractal principles into financial models fosters a deeper understanding of self-similarity and complexity within financial markets, even in the absence of empirical data.

Time Dependence of CAPM Betas on the Choice of Interval Frequency and Return Timeframes: Is There an Optimum?

The traditional CAPM beta is almost exclusively calculated over a return period that spans a window length of 60-months, at one-month return frequencies. It is one of the most utilized models in the asset management industry to assess systematic risk. Yet there is limited evidence to suggest that these estimation parameters are optimal. Utilizing data between January 2000 and December 2021 for the Russell 1000 index, we test daily, weekly, and monthly beta estimations to calculate tracking errors (TE) for the use of these betas in predicting subsequent performance over daily, weekly, and monthly timeframes. We identify that daily CAPM betas are best for predicting subsequent period daily returns and that weekly CAPM betas are strongly correlated with forward weekly and monthly period returns. Leveraging the significant advances in computing resources and the increasing utilization of high frequency trading strategies, we argue that additional window length and return interval-based CAPM betas should be calculated for estimating the systematic risk embedded in diversified portfolios.

Algorithmic Finance: Algorithmic Trading across Speculative Time-Spaces

The speeds at which transactions are completed in global financial markets are accelerating and, in the process, connecting financial centers around the globe like never before. Algorithmic trading at high frequency is a form of automated trading in which machines, rather than humans, make the decision to buy or sell in spatiotemporal sequences. Insofar as they have agency of their own, their actions support the owners of the means of production. These techniques codevelop with new financial geographies. Accordingly, I examine technological change and speculative time-spaces of algorithmic strategies at stock exchanges. By analyzing algorithmic finance, I examine how—and to what extent—time, speed, location, and distance become critical for algorithmic finance by configuring time-spaces as competitive factors. The analysis interprets time-spaces of high-frequency trading strategies through the ways in which algorithmic finance constititutes what I term mobile market-informational epicenters. This article discusses the spatiotemporalities of market information and examines whether space-times of privately owned high-frequency trading infrastructures result in a juxtaposition between “public” and “private” market information across digital and physical space. It thereby responds to the questions of what role geography plays when algorithms make money in microseconds and how techno-financial time-spaces turn into competitive advantage.

Does the Diversity and Solvency of Authorized Participants Matter for Bond ETF Arbitrage? Evidence from the Dash for Cash Episode

Authorized participants’ (APs) arbitrage in primary markets for ETF shares plays a key role in limiting dislocation in ETF prices. This paper builds a novel dataset of detailed US bond ETF-AP relationships and shows that high AP leverage played a significant role in weakening this arbitrage during the dash-for-cash episode of March 2020. The strength of the arbitrage relationship linking price signals to primary market activity weakened by 77 percent in ETFs related to more leveraged APs versus 64 percent for ETFs linked to less leveraged APs. This effect was particularly strong among those ETFs focusing on less liquid asset classes, relying on APs engaging in high-frequency trading strategies, and unrelated to banks and bank holding companies. Policy announcements by the Federal Reserve did not have had a strong impact in restoring arbitrage strength. AP leverage constraints operated in parallel to constraints faced by lead-market makers in secondary ETF markets, which were more closely related to regulatory capital limits.

High‐Frequency Trading and Financial Time‐Series Prediction with Spiking Neural Networks

A novel new method is introduced - “unsupervised spike learning” - which predicts spikes in price time series instead of price movement and direction. Three rewarding high-frequency trading strategies are developed and backtested by Kang Gao, Wayne Luk, and Stephen Weston.

Spoofing: effective market power building through perception alignment

PurposeThis paper aims to show that market power exists in financial markets and analyze how spoofing is used by a high-frequency trader to build market power by taking advantage of both behavioral weaknesses of individual investors and microstructural loopholes of trading venues.Design/methodology/approachAfter showing that market power exists in the financial market, this paper centers around the question of how market power is constructed in the financial market. To sharpen the answer to the question, the paper compares the conditions needed for market power construction in the financial market with those needed in the goods market. The paper selects spoofing, the frequently used order-based tactic in high-frequency trading strategies, to analyze in detail how spoof orders can ignite herding with market power building as the essence. The Flash Crash that occurred in the New York Stock Exchange on May 6, 2010 provides an excellent case of market power construction exhibited in spoofing.FindingsThe behavioral mechanism of market power construction in the case of spoofing is perception alignment. It becomes effective when two necessary conditions are met: the spoof trader takes advantage of the incomplete order display set up by the exchange; and the behavioral weaknesses exhibited by numerous individual investors. In addition to these key conditions, this paper finds other ones for market power to be created in the financial market. They are easier, quicker, more secret, more flexible and less risky relative to the conditions for market power building in the goods market.Practical implicationsThe detailed analysis points to the behavioral mechanism, i.e. perception alignment, and microstructural mechanism, i.e. incomplete order display, that could be responsive to regulation.Originality/valueThe originality of the findings is to uncover the mechanism of spoofing in taking advantage of behavioral biases of individual investors. The value is to gain more complete understanding of the essence of herding caused by spoofing.

The high frequency trade off between speed and sophistication

Central to the ability of a high frequency trader to make money is speed. In order to be first to trading opportunities, firms invest in the fastest hardware and the shortest connections between their machines and the markets. Yet this is not enough: algorithms must be short, no more than a few instructions. As a result there is a trade-off in the design of optimal high frequency trading strategies: being the fastest necessitates being less sophisticated. To understand the effect of this tension a computational model is presented that captures latency, both of code execution and information transmission. Trading algorithms are modelled through genetic programming with longer programmes allowing more sophisticated decisions at the cost of slower execution times. It is shown that, depending on the market composition, short fast strategies and slower more sophisticated strategies may both be viable and exploit different trading opportunities. The relative profits of these different approaches vary, however, slow traders benefit and social welfare increase in the presence of HFTs. A suite of regulations are tested to manage the risks associated with high frequency trading, the majority are found to be ineffective, though constraining the ratio of orders to trades may be promising.

Trading on Earthquakes: Algorithmic Financialization of Tectonic Events at Global Stock Exchanges

This paper addresses how high-frequency trading adapts to natural disasters. Stock market trades have accelerated to a rate at which shares change hands in microseconds. I examine ways in which high-frequency trading both reconfigures the dynamics of finance and changes the global financial system in different spatio-temporal ways and produces political ecologies of engagement, divergence and convergence between financial and Earth systems. Accordingly, I examine technological change and algorithmic strategies at stock exchanges. By analyzing algorithmic strategies, I investigate the connections between non-human trading and natural disasters. The analysis explains the nature of high-frequency trading strategies and market responses to three earthquakes in Japan. In the final section, I discuss how financial investment algorithms constitute a temporal informational epicenter when tectonic events are made subject to trading.

Trading on earthquakes – Algorithmic financialization of tectonic events at global stock exchanges

This paper addresses how high-frequency trading adapts to natural disasters. Stock market trades have accelerated to a rate at which shares change hands in microseconds. I examine ways in which high-frequency trading both reconfigures the dynamics of finance and changes the global financial system in different spatio-temporal ways and produces political ecologies of engagement, divergence and convergence between financial and Earth systems. Accordingly, I examine technological change and algorithmic strategies at stock exchanges. By analyzing algorithmic strategies, I investigate the connections between non-human trading and natural disasters. The analysis explains the nature of high-frequency trading strategies and market responses to three earthquakes in Japan. In the final section, I discuss how financial investment algorithms constitute a temporal informational epicenter when tectonic events are made subject to trading.

Big data analytics using multi-fractal wavelet leaders in high-frequency Bitcoin markets

Abstract We employ a time-scale multi-fractal decomposition approach to investigate the properties of Bitcoin prices and volume at different sampling rates using high-frequency data. We provide evidence of multi-fractality at all rates. The big data-driven analysis combined with statistical testing shows evidence of dominant multi-fractal traits within the intervals of 5 mn–90 mn, and 120 mn up to 720 mn. Wavelet leaders comprise a promising algorithmic technique that provides a richer description of the singularity spectrum. In particular, we reveal the distinct heterogeneity of the three log-cumulants for prices and volume between the two distinctive high-frequency sampling intervals. Our findings may assist in devising profitable high-frequency trading strategies in crypto-currency markets.

High-frequency trading: Order-based innovation or manipulation?

High-frequency trading (HFT) is a financial innovation that focuses on order flow and relies on quickly evolving information and communication technology. The innovation is successful, and HFT is highly and consistently profitable. However, the Flash Crash on 6 May 2010 exposed the unfamiliar side of HFT, thus illuminating the emergent need to unveil the negative impact that HFT has on other investors and the market. This paper examines data regarding quote-stuffing, spoofing, and market making provided by high-frequency (HF) traders, based on the increasing empirical literature. It first defines order-based manipulation (OBM) as the framework under which quote-stuffing, spoofing, and HF market making find common ground. It then provides details regarding how OBM is displayed in the three manipulation tactics. In essence, they all seek and exercise monopoly power in trading albeit through different ways of achieving it. The shared purpose is to gain monopolistic profit. The essence and common purpose explain why HF traders are not net liquidity providers, contrary to some proponents’ conclusions. Rather, this paper points out the three consequences that HF traders have brought to the market, i.e. increased volatility, increased frequency of unfairness, and instability potential. Recent regulatory improvement and completed prosecutions against manipulative HFT strategies justify the analysis.

Grid Trading System Robot (GTSbot): A Novel Mathematical Algorithm for Trading FX Market

Grid algorithmic trading has become quite popular among traders because it shows several advantages with respect to similar approaches. Basically, a grid trading strategy is a method that seeks to make profit on the market movements of the underlying financial instrument by positioning buy and sell orders properly time-spaced (grid distance). The main advantage of the grid trading strategy is the financial sustainability of the algorithm because it provides a robust way to mediate losses in financial transactions even though this also means very complicated trades management algorithm. For these reasons, grid trading is certainly one of the best approaches to be used in high frequency trading (HFT) strategies. Due to the high level of unpredictability of the financial markets, many investment funds and institutional traders are opting for the HFT (high frequency trading) systems, which allow them to obtain high performance due to the large number of financial transactions executed in the short-term timeframe. The combination of HFT strategies with the use of machine learning methods for the financial time series forecast, has significantly improved the capability and overall performance of the modern automated trading systems. Taking this into account, the authors propose an automatic HFT grid trading system that operates in the FOREX (foreign exchange) market. The performance of the proposed algorithm together with the reduced drawdown confirmed the effectiveness and robustness of the proposed approach.

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

ABSTRACTMarket 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.

Legal Liability for Fraud in the Evolving Architecture of Securities Markets: From Marketplaces to Traders

This chapter analyses the reach of legal liability for securities fraud in the United States in relation to certain types of abuses often associated with high-frequency trading strategies. The nature of algorithmic and high-frequency trading presents new challenges for establishing legal liability for fraud in securities markets. It specifically examines the reach of legal liability in relation to two, sometimes overlapping, entities: marketplaces and proprietary traders. Through an examination of various informational asymmetries in securities markets, including the order type controversy, this chapter finds that establishing liability for securities fraud is significantly murkier for proprietary traders than for marketplaces. Subsequently, this study addresses potential approaches to the reach of liability to traders and discusses mechanics and proper characterization of the underlying harm.

High‐frequency trading from an evolutionary perspective: Financial markets as adaptive systems

AbstractThe recent rapid growth of algorithmic high‐frequency trading strategies makes it a very interesting time to revisit the long‐standing debates about the efficiency of stock prices and the best way to model the actions of market participants. To evaluate the evolution of stock price predictability at the millisecond timeframe and to examine whether it is consistent with the newly formed adaptive market hypothesis, we develop three artificial stock markets using a strongly typed genetic programming (STGP) trading algorithm. We simulate real‐life trading by applying STGP to millisecond data of the three highest capitalized stocks: Apple, Exxon Mobil, and Google and observe that profit opportunities at the millisecond time frame are better modelled through an evolutionary process involving natural selection, adaptation, learning, and dynamic evolution than by using conventional analytical techniques. We use combinations of forecasting techniques as benchmarks to demonstrate that different heuristics enable artificial traders to be ecologically rational, making adaptive decisions that combine forecasting accuracy with speed.