High-Frequency Trading (HFT) represents a significant evolution in the financial markets, leveraging advanced technologies to execute large volumes of trades in mere fractions of a second. Born from the amalgamation of sophisticated algorithms, lightning-fast computer networks, and intricate strategies, HFT has redefined the landscape of trading in modern times.

HFT's prowess is reflected in its ability to identify minute inefficiencies in the market and capitalize on them swiftly. These trading opportunities, which may last only milliseconds, are the result of minor price discrepancies across trading venues or small informational advantages. For instance, HFT strategies might exploit the split-second delays in stock price updates between exchanges.

The significance of HFT in today's trading world cannot be understated. As of the early 2020s, HFT accounted for approximately half of the equity market volume in the U.S. Its impact is pervasive, influencing everything from market liquidity to the pricing of securities. Proponents praise HFT for reducing trading costs, improving market efficiency, and enhancing liquidity. Meanwhile, skeptics raise concerns about potential market instability or the fairness of allowing certain traders an advanced technological edge.

The financial markets have always been arenas of innovation. In their early days, exchanges were characterized by open-outcry systems, where traders would shout and use hand signals to communicate orders. Floor traders had to be sharp, reacting to market movements in real-time. It was a physical, chaotic, yet organized dance of supply, demand, and human intuition.

Fast forward to the late 20th century, and the transformation began with the digitalization of stock markets. Electronic trading platforms started replacing the cacophony of trading floors. While these electronic systems initially just digitalized existing manual processes, they opened the door for the next revolutionary step: algorithmic trading.

By the early 2000s, algorithmic trading had made its mark. Trading decisions, which were once rooted in human judgment, were now driven by mathematical models and executed by computers. These algorithms could analyze vast datasets and execute orders based on predefined criteria at speeds far beyond human capability. Not only did this increase the speed and precision of trades, but it also minimized the emotional biases inherent to human decision-making.

With the continual advancements in technology, a new subset of algorithmic trading began to emerge: high-frequency trading (HFT). HFT utilizes extremely sophisticated algorithms to trade thousands of orders in fractions of a second. By the 2010s, HFT had exploded in popularity, accounting for a substantial proportion of total trading volume in many global markets[1]. The key to HFT's success lies in its speed. Propelled by technological breakthroughs in processing power, data transmission, and software efficiency, HFT firms sought every possible advantage to shave microseconds off their execution times.

This relentless drive for speed led to innovations like co-location, where trading firms placed their servers physically close to exchange servers to reduce signal travel time. Meanwhile, advances in telecommunications allowed for faster data transmission, with companies even laying straighter transatlantic cables to trim transmission time[2].

From shouting orders across a bustling trading floor to executing millions of trades in the blink of an eye, the journey of high-frequency trading underscores the financial market's ability to evolve. With each technological leap, trading strategies and the market's very structure adapted, leading to the hyper-efficient, electronically driven landscape we see today.

Use our strategy database to develop quantitative strategies faster.

High-Frequency Trading (HFT) is more than just algorithms making rapid-fire trades. It's a complex ecosystem that leverages cutting-edge technology at every juncture to ensure maximum efficiency and speed. Let's dissect this intricate machinery.

Infrastructure: Hardware, Software, and Data

The foundation of HFT is its robust infrastructure. Custom-built servers, tailored for trading, house the algorithms. These servers are equipped with the latest processors to facilitate quick decision-making and trade execution. The software, an equally crucial component, is meticulously optimized for performance. It undergoes rigorous testing, ensuring that algorithms can execute without hitches and adapt to rapidly changing market conditions.

Data is the lifeblood of HFT. Real-time market data feeds, often provided by exchanges or third-party vendors, stream into HFT systems. This data, consisting of price, volume, and order information, is continuously processed and analyzed.

Speed, Latency, and Co-location

In the world of HFT, every microsecond counts. Speed is paramount. However, while having the fastest algorithm is essential, ensuring that your trading command reaches the exchange swiftly is equally crucial. This is where latency comes into play. Latency refers to the time it takes for data to travel from the trader's system to the exchange. Reducing this delay is a never-ending quest in HFT.

Co-location is a solution to the latency issue. By positioning their servers within or extremely close to an exchange's data center, HFT firms can dramatically reduce the time it takes for their trade orders to reach the exchange. This proximity ensures that they're among the first to receive market data and can act upon it instantaneously.

Tick by Tick Data and Its Analysis

While many traders might look at aggregated data over seconds, minutes, or even days, HFT firms analyze data on a tick-by-tick basis. A "tick" is any change in the price of a security, regardless of volume. By examining every single price change, HFT algorithms can detect patterns, anomalies, or opportunities that might only exist for fractions of a second[3].

This granular data analysis demands immense computational power. Sophisticated algorithms sift through millions of data points every second, deciding when to buy or sell. Machine learning and other advanced techniques help in refining these strategies, allowing them to evolve with the market.

High-Frequency Trading is an intricate ballet of technology, data, and strategy. It's a world where millisecond improvements can mean the difference between profit and loss, and where the relentless pursuit of efficiency constantly reshapes the trading landscape.

Market making is a fundamental trading strategy not only in High-Frequency Trading (HFT) but across various financial markets. At its core, market making involves simultaneously quoting both buy (bid) and sell (ask) prices for a financial instrument, with the aim of profiting from the bid-ask spread.

Traditionally, market makers facilitated liquidity, ensuring that traders could enter or exit positions even in less actively traded instruments. They committed capital to buy securities when no one else was buying and to sell when no one else was selling.

In the realm of HFT, market making has taken a sophisticated turn. Advanced algorithms are employed to adjust bid and ask prices in real-time, responding to market dynamics instantaneously. The primary objective remains: capture the spread between buying and selling prices while minimizing inventory risk. The difference here is the speed and scale. HFT market makers operate at an incredibly high frequency, often making thousands of trades per second.

Real-world examples:

In essence, market making in the high-frequency domain is about providing liquidity at unparalleled speed, ensuring markets remain efficient even in turbulent times.

Statistical arbitrage, often shortened to StatArb, is a quantitative strategy that seeks to exploit temporary price anomalies between related securities. The strategy uses mathematical models and computations to identify trading opportunities with a theoretically predefined risk-return profile. Unlike traditional arbitrage strategies, which seek to exploit price discrepancies for the same asset across different markets, StatArb is based on the premise that there exists a stable, long-term relationship between the prices of certain assets, and when a deviation occurs, it provides a tradeable opportunity.

Types:

Real-world examples:

Statistical arbitrage, while based on rigorous mathematical foundations, is not risk-free. Traders need to account for transaction costs, potential model inaccuracies, and evolving market dynamics. Yet, with a disciplined approach and meticulous backtesting, StatArb can offer a compelling avenue for systematic, data-driven profits in the market.

Event arbitrage is a strategy centered around capitalizing on price movements that are anticipated to occur from specific events, like mergers, earnings announcements, or economic data releases. In the world of high-frequency trading, the quick dissemination and interpretation of these events can lead to significant trading advantages.

HFT firms leverage state-of-the-art technologies to instantaneously process and act upon news and announcements. When an announcement is made, algorithms can parse the news, analyze the potential market impact, and execute trades within milliseconds, often before most human traders can even read the headline.

How HFT Firms Utilize News and Announcements:

Tools Used for News-Based Trading:

The allure of event arbitrage in high-frequency trading lies in its potential for high returns. However, it's a double-edged sword. Given the speed and automation involved, if an algorithm misinterprets news or reacts inappropriately, it can lead to significant losses in a very short time frame.

Latency arbitrage is one of the most controversial and prevalent strategies in the high-frequency trading (HFT) world. This strategy thrives on exploiting minuscule, often microsecond, differences in the speed at which market participants receive and act on market information.

The significance of microseconds in latency arbitrage cannot be overstated. A mere microsecond advantage can lead to substantial profits for HFT firms. For instance, if one trader receives information about a large buy order a fraction of a second before others, they can buy the asset in question and then sell it moments later at a slightly higher price to the trader placing the large order. While the profit from each such transaction might be small, the sheer volume of these trades means that profits can accumulate rapidly.

To achieve and maintain these minuscule advantages, HFT firms employ a combination of advanced hardware, software, and strategic positioning. Here's how they minimize latency:

It's worth noting that while latency arbitrage can be lucrative, it's also fraught with challenges. The arms race for speed is relentless, requiring constant investment. Moreover, it has been a subject of regulatory scrutiny due to concerns about market fairness and stability.

Momentum strategies, at their core, are based on a simple premise: assets that have performed well in the recent past are likely to continue performing well in the near future, and conversely, assets that have performed poorly will continue their downward trajectory. High-frequency trading (HFT) leverages this principle but on a much shorter time scale, sometimes holding positions for mere seconds or even milliseconds.

To successfully implement momentum strategies in the HFT realm, traders employ sophisticated algorithms that are designed to detect the early signs of an emerging trend. These algorithms constantly analyze market data, seeking patterns, volume shifts, and price movements indicative of a potential trend. Once detected, the algorithm will initiate trades to capitalize on this momentum before it dissipates.

Several factors are critical for HFT momentum strategies:

HFT momentum strategies can be particularly effective during times of market volatility, where price movements are more pronounced. However, they also come with challenges. The intense competition among HFT firms means that the first to detect and act on a trend often reaps the most benefit. Latecomers might find diminished returns or even losses if they enter just as the momentum is reversing.

While some critics argue that HFT momentum strategies can exacerbate market volatility, proponents believe that these strategies provide additional liquidity and can help "correct" mispriced assets more rapidly.

Order flow prediction revolves around anticipating large trades before they execute and gauging the subsequent market impact. Successfully predicting order flow allows traders to position themselves advantageously, potentially benefiting from the anticipated price movements.

In the realm of high-frequency trading (HFT), the stakes are high, and the margins thin. The ability to predict large orders can be a significant advantage. Here's how it works:

Critics argue that trading based on order flow prediction can be parasitic, with HFTs profiting at the expense of institutional and retail traders. They contend that it could lead to predatory trading practices where HFTs exploit their speed advantage to front-run large orders. On the flip side, proponents claim that HFT adds liquidity to the market and ensures tighter bid-ask spreads.

In the fast-paced environment of high-frequency trading, detecting liquidity is like finding hidden treasure. Liquidity, in trading terms, refers to the ease with which assets can be bought or sold without causing significant price movements. In markets where liquidity is high, substantial trades can be executed without drastically affecting prices. Conversely, in less liquid markets, even small trades can lead to considerable price shifts.

For HFT traders, spotting liquidity (especially hidden liquidity) provides strategic advantages, enabling more informed decision-making and risk management. Hidden orders and large trades are prime sources of this concealed liquidity. Here's how HFT systems approach liquidity detection:

Detecting liquidity is a double-edged sword. On one hand, it can provide HFT traders with opportunities to achieve better execution prices and reduce trading costs. On the other hand, if misused, it can be seen as predatory, with HFTs potentially taking advantage of other market participants. This balance, and the methods employed to strike it, is a subject of ongoing debate within the trading community.

Algorithmic execution strategies are the backbone of modern trading, ensuring large orders are managed effectively to reduce adverse market impact. When a substantial trade is executed in one go, it can significantly move the market, leading to slippage and, consequently, a less favorable execution price. To circumvent this, traders use algorithmic execution strategies to break down large orders into smaller, manageable chunks, executed over time or across different venues.

Algorithmic execution strategies are not a one-size-fits-all solution. Their efficiency depends on factors like order size, market volatility, and the specific objectives of the trader. However, their primary aim remains consistent: optimizing trade execution by balancing urgency, cost, and market impact to achieve the best possible result.

High-frequency trading has always thrived on the cutting edge of technology, and today, Artificial Intelligence (AI) and Quantum Computing are among the pioneering technologies leading the charge.

AI's role in HFT primarily revolves around machine learning and deep learning, branches that equip algorithms to process vast amounts of data, identify patterns, and make decisions at lightning speeds. These AI algorithms are tailor-made to predict minute market movements, analyze order flow, and adjust strategies in real-time. For example, convolutional neural networks, which were originally designed for image processing, are now being utilized to analyze patterns in high-dimensional trading data[7]. Reinforcement learning, another subset of AI, enables algorithms to learn optimal trading strategies by interacting with the market, akin to learning a complex game. Through iterative processes, these algorithms can refine their strategies to maximize profits.

Moreover, AI is making strides in sentiment analysis. By analyzing vast amounts of unstructured data from news articles, financial reports, and social media, AI algorithms can gauge market sentiment and predict short-term price movements. Such capabilities are invaluable in HFT, where even the smallest price prediction edge can translate into significant gains.

On the frontier of computational power, quantum computing promises to revolutionize HFT. Traditional computers use bits as the smallest unit of data, which can be either 0 or 1. Quantum computers, on the other hand, use quantum bits (qubits), which can exist in a superposition of states, allowing them to process a massive amount of information simultaneously[8]. In HFT, where latency is paramount, quantum computers have the potential to calculate optimal trading strategies in fractions of the time taken by classical computers.

Beyond strategy formulation, quantum computing can play a pivotal role in risk analysis and portfolio optimization. Quantum algorithms, like the quantum amplitude estimation, can be used to compute risk measures much faster than classical algorithms, providing HFT firms a considerable advantage in risk-adjusted returns[9].

However, it's essential to understand that quantum computing in HFT is still in its infancy. While there are promising developments, scalable and error-free quantum computers that outperform classical computers are yet to be achieved. But the potential is undeniable, and as the technology matures, it could redefine the boundaries of HFT.

In essence, both AI and quantum computing are pushing the envelope of what's possible in HFT. While AI algorithms are already driving significant change, delivering more accurate predictions and dynamic strategies, quantum computing holds the promise of unmatched computational power, paving the way for an entirely new era of trading.

High-Frequency Trading (HFT) has undeniably left a significant footprint on global financial markets. As with any powerful force, it brings with it a blend of benefits and drawbacks, often sparking intense debates among market participants, regulators, and the public.

On the positive side, proponents argue that HFT enhances market quality by providing several key benefits:

However, HFT's rise hasn't been without its controversies and criticisms. Some of the most cited concerns include:

Beyond specific practices, there's an overarching ethical and moral debate around HFT. Critics question whether it's appropriate for traders to profit simply from being faster, without necessarily adding substantive value to the market. Additionally, the vast resources spent on gaining minuscule speed advantages—like setting up co-location facilities close to exchange servers—might be seen as a wasteful "arms race".

Defenders argue that HFT is merely the natural evolution of trading in a technologically advanced world and that the overall benefits to market quality outweigh the drawbacks. They contend that many criticisms stem from misunderstandings or are issues that can be addressed through proper regulation rather than vilifying the entire practice.

Regardless of one's stance, it's clear that HFT remains one of the most polarizing topics in modern finance. Its profound influence on market dynamics ensures that discussions around its role, impact, and ethics will remain at the forefront for years to come.

High-Frequency Trading (HFT) operates within a complex regulatory landscape that varies across different jurisdictions. This landscape reflects a balance between fostering innovation and ensuring market integrity, protecting investors, and preventing abuse.

In the U.S., the Securities and Exchange Commission (SEC) and the Commodity Futures Trading Commission (CFTC) oversee HFT activities. Following the 2010 Flash Crash, the regulators took steps to increase oversight and control over automated trading. Some notable regulatory measures include:

Across the Atlantic, Europe has its regulatory framework predominantly shaped by the Markets in Financial Instruments Directive II (MiFID II). Introduced in 2018, MiFID II addresses several aspects of HFT:

As markets and technologies evolve, so do regulations. Potential future regulations may focus on:

Given the global nature of financial markets, differences in regulations across countries are crucial for HFT firms operating internationally. For example, Asian markets like Japan and Hong Kong have unique regulatory stipulations tailored to their market structures. Firms must navigate this patchwork of rules, ensuring compliance while optimizing their strategies.

To conclude, the regulatory landscape for HFT is continually evolving, shaped by market events, technological advancements, and shifts in regulatory philosophy. While it's challenging to predict the exact future trajectory, the overarching goal will remain: ensuring that financial markets are fair, transparent, and resilient.

High-Frequency Trading (HFT) is dominated by firms that can invest significantly in technology, infrastructure, and talent. Let's shed light on the major players in this space and how retail traders fit into the HFT ecosystem.

Globally, several firms are renowned for their prowess in HFT. Virtu Financial is one such powerhouse, known for its advanced trading algorithms and vast global reach[15]. Citadel Securities, a part of Citadel LLC, is another behemoth that handles a significant portion of U.S. equities trading volume[16]. Two Sigma Securities, Flow Traders, and Hudson River Trading are also formidable entities in the HFT world, known for their sophisticated trading strategies and tech-forward approaches. These firms, among others, play critical roles in providing liquidity, reducing trading costs, and contributing to efficient market operations.

Shifting focus to India, the landscape is a blend of domestic and international firms. Traditional brokerages, such as Zerodha and Sharekhan, have expanded their technological capabilities to offer algo-trading solutions, though not purely in the HFT category. International entities like Optiver and Tower Research Capital have also established their presence, leveraging the country's tech talent pool and growing financial markets[17].

The role of retail traders in HFT is nuanced. Direct participation in pure HFT is challenging for retail traders due to the enormous capital requirements, sophisticated technology, and lightning-fast trade execution speeds. However, retail traders do engage in algorithmic trading, a broader category under which HFT falls. With the democratization of technology, platforms like QuantConnect and Alpaca have made it easier for retail traders to develop and deploy algorithmic strategies[18]. Though they might not compete directly with HFT giants, they bring diversity to market participation and benefit from the tighter spreads and increased liquidity that HFT provides.

To understand HFT's ecosystem, recognizing its major players and the interplay between institutional and retail participants is essential. As technology continues to evolve and markets grow, the dynamics between these entities might shift, but their collective contribution to modern trading cannot be understated.

High-Frequency Trading (HFT) is a tech-driven domain, where the smallest of time advantages can result in significant profits or losses. As such, the choice of hardware, software, and infrastructure is paramount to an HFT firm's success.

In the realm of hardware, the primary requirement is to achieve minimal latency. Specialized high-speed servers are a given. These servers often employ FPGA (Field Programmable Gate Array) technology, allowing for hardware-level customization of trading algorithms, granting traders a speed advantage. HFT setups also invest in high-speed network equipment like 10 Gigabit Ethernet switches and dedicated fiber-optic lines to ensure optimal connectivity.

Co-location is another critical aspect. Many HFT firms place their servers in the same data centers as exchanges, reducing the physical distance data needs to travel. This co-location allows for faster order execution and data receipt.

On the software front, HFT systems require a robust and highly efficient trading platform. This software needs to handle vast amounts of data at lightning speeds. Some firms opt for custom-built platforms tailored to their unique trading strategies. Others might choose commercially available platforms that offer high customization levels. The software's primary functions include market data processing, algorithm execution, risk management, and order routing.

Cost considerations in HFT are significant. The initial setup, involving hardware, software, and co-location services, can run into millions of dollars. Add to this the ongoing costs of data feeds, system maintenance, and periodic upgrades. It's a capital-intensive endeavor, with the upfront costs acting as a barrier to entry for many potential participants.

However, the potential returns are also substantial. While the profit per trade might be minuscule due to the tight bid-ask spreads, the volume of trades executed daily can result in considerable aggregate profits. Efficient HFT setups, combined with the right strategies, can achieve annual returns significantly higher than traditional trading methods.

In summary, HFT infrastructure is about speed, efficiency, and reliability. The right setup, though costly, can be a linchpin for success in the competitive world of high-frequency trading.

The landscape of High-Frequency Trading (HFT) is rapidly evolving, influenced by technological innovations, regulatory changes, and market dynamics. As the industry ventures into the future, it is poised to face both challenges and opportunities that could shape its trajectory.

One of the primary challenges HFT faces is the constant race for speed. As firms invest heavily in achieving minimal latency, the diminishing returns from incremental speed improvements become evident. The difference between microseconds and nanoseconds might not always translate into proportional profit increases, making the cost-to-benefit ratio of such advancements questionable.

Regulation also presents a challenge. Concerns over market stability, transparency, and fairness have prompted regulatory bodies worldwide to reconsider the oversight of HFT activities. Stricter rules could impede the flexibility of HFT strategies, potentially leveling the playing field but at the cost of some advantages HFT currently enjoys[19].

Conversely, the rise of alternative trading venues and decentralized finance (DeFi) platforms offers new arenas for HFT to operate. These platforms might bring about different market dynamics, presenting opportunities for innovative trading strategies that exploit the unique characteristics of these venues.

Moreover, the fusion of advanced machine learning models with HFT might unlock new dimensions in trading. While speed will always be paramount, the ability to predict market movements with greater accuracy using AI can be a game-changer, offering a competitive edge even if speed advantages plateau.

Comparing HFT with traditional investment strategies sheds light on their distinct objectives and time horizons. While HFT focuses on exploiting minute price discrepancies within short time frames, traditional strategies are more concerned with the long-term valuation of assets based on fundamentals. HFT seeks quick profits from fleeting opportunities, whereas traditional investment might involve holding assets for months or years, banking on their intrinsic value growth.

In essence, HFT and traditional investment operate on different paradigms. However, as markets evolve, a potential convergence or symbiosis might emerge. For instance, long-term investors can leverage insights from HFT patterns to understand market microstructures, while HFT firms might utilize long-term trends to inform some of their strategies.

In conclusion, the future of HFT is a blend of challenges and potential, shaped by technological advancements, regulatory landscapes, and market evolutions. Its interplay with traditional investment methods underscores the vast diversity and complexity of the financial world.

In essence, High-Frequency Trading stands as a testament to the fusion of finance and technology. Its intricate strategies, while controversial to some, undeniably play a pivotal role in today's dynamic trading world.

High-Frequency Trading (HFT) has evolved from the nascent stages of manual trading to become a dominating force in the global financial markets. Rooted in technological advancements, HFT exemplifies the synergy between finance and technology, ensuring ultra-fast trade executions and exploiting minute price discrepancies. As strategies like Market Making and Latency Arbitrage have matured, newer paradigms such as the incorporation of AI and Quantum Computing are on the horizon, promising unprecedented efficiencies.

Yet, the world of HFT is not without its complexities and challenges. Controversies such as flash crashes have prompted rigorous debates on the ethical implications and the overall stability of the financial system in the age of HFT. Regulatory landscapes across the globe are in flux, with authorities attempting to strike a balance between fostering innovation and ensuring market fairness and stability.

The essence of HFT lies not just in its speed but in its ability to redefine market dynamics continually. As we look to the future, while HFT may face challenges, its integral role in shaping the financial markets remains undeniable. For traders, investors, and market enthusiasts, understanding HFT is paramount in navigating the evolving world of finance and staying ahead in the game.

💡 Read more:

[1]: Hendershott, T., Jones, C. M., & Menkveld, A. J. (2011). Does algorithmic trading improve liquidity?. The Journal of Finance, 66(1), 1-33.

[2]: Aldridge, I. (2013). High-frequency trading: A practical guide to algorithmic strategies and trading systems. John Wiley & Sons.

[3]: Kirilenko, A. A., Kyle, A. S., Samadi, M., & Tuzun, T. (2017). The flash crash: High-frequency trading in an electronic market. The Journal of Finance, 72(3), 967-998.

[4]: Kirilenko, A., & Lo, A. W. (2013). Moore's law versus Murphy's law: Algorithmic trading and its discontents. Journal of Economic Perspectives, 27(2), 51-72.

[5]: Avellaneda, M., & Lee, J. H. (2010). Statistical arbitrage in the U.S. equities market. Quantitative Finance, 10(7), 761-782.

[6]: Almgren, R., & Chriss, N. (2005). Optimal execution of portfolio transactions. Journal of Risk, 3, 5-39.

[7]: Dixon, M. F., Klabjan, D., & Bang, J. H. (2019). Classification-based financial markets prediction using deep neural networks. Algorithms, 12(8), 156.

[8]: Montanaro, A. (2016). Quantum algorithms: an overview. npj Quantum Information, 2(1), 1-6.

[9]: Woerner, S., & Egger, D. J. (2018). Quantum risk analysis. npj Quantum Information, 5(1), 1-10.

[10]: Brogaard, J., Hendershott, T., & Riordan, R. (2014). High-frequency trading and price discovery. The Review of Financial Studies, 27(8), 2267-2306.

[11]: Hasbrouck, J., & Saar, G. (2013). Low-latency trading. Journal of Financial Markets, 16(4), 646-679.

[12]: Hendershott, T., Jones, C. M., & Menkveld, A. J. (2011). Does algorithmic trading improve liquidity? The Journal of Finance, 66(1), 1-33.

[13]: Gomber, P., Arndt, B., Lutat, M., & Uhle, T. (2013). High-frequency trading.

[14]: Kirilenko, A. A., Kyle, A. S., Samadi, M., & Tuzun, T. (2017). The flash crash: High-frequency trading in an electronic market. The Journal of Finance, 72(3), 967-998.

[15]: Virtu Financial - About Us.

[16]: Citadel Securities - What We Do.

[17]: Optiver - About Us.

[18]: QuantConnect - About Us.

[19]: Regulating High-Frequency Trading