High-frequency trading is a form of automated trading that uses sophisticated computer technology and systems to execute trades at millisecond speeds and hold positions for short periods of time during the day. Among these, liquidity trading strategies, market microstructure trading strategies, event trading strategies and statistical arbitrage strategies are popular in mature markets abroad.
High-frequency trading is a rising star of the financial market and a crystal of financial and technological developments. The rapid development of high-frequency trading has attracted great market interest in recent years. High-frequency trading has been lacking a strict definition, citing the definition of the European Securities and Markets Authority: high-frequency trading is a form of automated trading, which uses complex computer technologies and systems to execute trades at millisecond speeds and hold stocks for short periods of time during the day.
High-frequency trading has several key characteristics: processing transaction data, algorithmic trading, high cash flow rate and intraday trading. Processing transaction data and algorithmic trading is an important process of high-frequency trading. High-frequency trading collects, processes and analyzes market data on micro-trading opportunities by collecting, processing and analyzing market data. High-frequency trading has four types of trading strategies that are popular in mature foreign markets.
The liquidity trading strategy
A liquidity trading strategy is a trading strategy that provides liquidity to the market in order to make a profit. Market makers provide the market with a book of orders at different price levels to provide liquidity to the position recipients, so it is called a liquidity trading strategy. Market makers contribute to the liquidity of the market, and many non-active markets have significantly increased liquidity and significantly reduced trading costs due to the presence of market makers. The theoretical basis of the market maker strategy is the stock and information model. The stock model was proposed by Demsetz in 1968 in his book, The Cost of Trading in the Binary Options Market. He argues that the bid-ask spread is actually a compensation provided by an organized market for the immediacy of the transaction. The information model was proposed by Bagehot in 1971. He argues that the bid-ask spread is caused by the asymmetry of market information.
Micro-structured trading strategies in the market
The market microstructure trading strategy is mainly based on the analysis of the market's instantaneous discount data, the strategy of ultra-short trades based on the imbalance of the buy and sell order flow in the short term. There are many trading opportunities hidden in the market's instantaneous buy and sell order flow, by observing the visible order book condition, analyze whether the sell one-stream dominates or buy one-stream dominates in the very short term. Market microstructures allow traders to compare the strengths of buy and sell orders in the order book, to pre-trade, and to quickly settle. The premise here is that the information in the order book is genuinely representative of the investor's intentions, but in fact the order book information can also be interfered with. It is worth mentioning here that the brokers in the domestic futures trading, their trading strategy belongs to this category, that is, by observing the changes in the order flows in the market, looking for trading opportunities, quick manual ordering. The brokers have a small amount of money in the market, but the volume of transactions produced is very large, can enter and exit the market hundreds of times a day, the profitability of the good brokers and the capital curve is very amazing. Such trading strategies require a high level of human responsiveness, which can stand out from the crowd. We learned from our counterparts in the Taiwanese futures industry that in Taiwan, man-made high-frequency trading has been completely defeated by computer-automated high-frequency trading.
Trading strategies for events
Event trading strategy is a strategy to trade using the market's response to events. Events can be events that affect a wide range of economic events, or they can be industry-related events. Each event has a large time difference in its impact on the market. There are two key elements to this strategy. The first is to determine what events can have an impact. The question may seem strange, but experienced traders know that the impact of events on the market is complex, and a perfectly beneficial event can have the opposite effect in different sectors and time windows. The market also has an expectation of events, many of which do not happen, and the market has already had a predicted reaction, so it is necessary to first determine what events constitute an expected change.
The statistical strategies
A statistical arbitrage strategy is a trading strategy that searches for securities with a long-term statistical relationship and uses them to leverage when the price difference between the two deviates. The statistical arbitrage strategy is widely used in all types of securities markets, including stocks, futures, foreign exchange, etc. The famous American long-term capital management company (LTCM) is a hedge fund company based on statistical leverage. LTCM once created a dismal performance, starting with a net asset value of $1.25 billion, and by the end of 1997, it rose to $4.8 billion, with a net growth of 2.84 times. Annual return on investment was 28.5% in 1994, 42.8% in 1995, 40.8% in 1996, and 17% in 1997. Unfortunately, the Russian financial storm shattered his myth, with net assets falling by 90% in just 150 days, losses of $4.3 billion, and on the verge of bankruptcy. This also tells us that no matter how good the statistical model is, there are limits and risk control should always come first.