What is programmed trading? What is quantitative investment? In fact, most of us did not try to understand the correct concepts of programmed trading and quantitative investment, but drew an equal sign between them in general. There are even so-called commentators who directly confuse high-frequency trading with low-delay trading, pointing out that deer is a horse and misleading public opinion.
In the earliest definition of new york Stock Exchange (NYSE), programmed trading refers to a basket of transactions with more than 15 stocks and a turnover of more than10 million US dollars. In August, 2065,438+03, NYSE cancelled the condition that the total value in the definition exceeded 654.38+0 million dollars. In the later market practice, the object of programmed trading was extended to stocks, futures and options listed on various exchanges, and the instructions were directly issued by computers and automatically executed.
With the continuous enrichment of financial derivatives, programmatic trading has become an important means of transaction realization for institutional investment, which has solved the problems of placing orders in batches, uninterrupted trading in the whole market and reducing the impact cost. This is a technical means, using software to place orders instead of manual entrustment. Institutions adopt programmed trading means in the hope of accelerating the formation of prices while minimizing market impact. For example, algorithmic transactions such as VVAP and TWAP are very basic programmatic transactions, and they will not amplify the ups and downs themselves.
The concept of quantitative investment is broader. Usually, the systematic investment method that we understand, with data model as the core and programmatic trading as the means, can automatically identify investment opportunities and automatically trigger transactions, is quantitative investment. It is not difficult to see that quantitative investment is a routine of investment, and programmatic trading is only an objective means of transaction realization. To the extreme, compared with on-site bidding, the process of investors reporting to the front-end machine of securities firms through various software and then entering the exchange for transaction is actually a programmed transaction.
We can't simply understand quantitative investment as programmed trading, let alone programmed trading as high-frequency trading. Quantization is Tao, and programming is art, which cannot be confused. Quantitative investment has great development potential in China. Although a series of measures to curb excessive speculation in China will involve quantitative investment, its starting point is still aimed at procedural trading behaviors that may disrupt market order, so as to maintain market fairness. Specific methods can be discussed, but clear concepts and types are indispensable prerequisites for regulators and practitioners.
Programmed transaction grows in controversy
In many periods of financial history, programmed transactions were associated with severe market fluctuations. Many countries have taken phased measures to curb programmed trading, but at the same time, many research results show that programmed trading has not destroyed the stability of the market. A large number of empirical studies show that programmed trading is not necessarily related to market price fluctuation, and there is no evidence that index arbitrage has aggravated market price fluctuation.
There has never been a shortage of devil traders in the history of financial transactions, and fat finger incidents can be said to be everywhere, but why does programmatic trading attract special attention from the market? The author believes that this is closely related to the fact that once there is a problem with programmed trading, it may lead to very serious consequences in a very short time and may cause a chain reaction in the whole market. Coupled with the "black box" characteristics of programmed transactions, after media propaganda and fermentation, the whole process is more confusing and newsworthy than the simple fat finger incident.
In fact, in several well-known vicious incidents in the global financial market, the wrong programmed trading is no more amazing in destructive power than the devil trader.
2065438+June 2005, a trader in the London foreign exchange department of Deutsche Bank mistakenly took the "net value" in an order as the "total amount", which made Deutsche Bank send $6 billion to an American hedge fund customer in vain. After urgent consultation, Deutsche Bank recovered the money the next day, and the shocked Deutsche Bank immediately made an executive adjustment.
Compared with the luck of Deutsche Bank this time, in May of 20 10, a fat finger incident of J-COM Mizuho Securities traders led to the final economic loss of Mizuho Securities exceeding 30 billion yen.
Another widely circulated event was that in June 2005, a broker of Fubon Securities Company mistook the English figure of 80 million yuan for 8 billion yuan when accepting a client's transaction entrustment. Results On that day, between 1 1: 33 and1:40, the Taiwan Province stock market index soared from 6,284.82 to 6,342.45, and more than 100 stocks had daily limit.
Among the iconic programmed trading accidents, the first two were the failure of Knight Capital's trading and the 20 10 US stock crash caused by Sarao's programmed trading.
On August 65438, 2065438, Knight Capital, the largest market maker in the United States at that time and the leading institution of high-frequency trading, made a mistake when updating the trading system of the market-making department the day before, and sent many wrong NYSE stock quotations to the market, resulting in abnormal fluctuations in the prices of more than 150 stocks, with a huge loss of $440 million in less than an hour. The direct consequence of the incident was that Knight Capital, which once enjoyed unlimited scenery, was acquired by competitors to avoid bankruptcy.
20 10 On May 6th, the US stock market plunged 9% mysteriously, and the Dow fell nearly 1000 points in a few minutes. The long investigation lasted for five years until Navinder Singh Sarao, a 37-year-old trader, was arrested in the UK in 20 15. The US Department of Justice said in a statement that the United States is seeking extradition. According to the disclosure, on the day of the U.S. stock market crash, Salao manipulated the S&P 500E-Mini index through grading algorithm and spoof algorithm, earning 900,000 U.S. dollars, while American investors lost nearly 1 trillion U.S. dollars due to the crash in a few minutes.
In addition, in recent years, many major exchanges around the world, including the Spanish stock exchange IBEX, Tokyo Stock Exchange and CME, have also experienced accidents of suspension of trading due to system problems. Since the programmed transaction was gradually recognized by the market, the regulatory agencies in various countries have never stopped to restrict and manage its operation, including the investigation of civil liability for those responsible for major events that affect market stability.
In short, it is not the programmatic trading technology itself that destroys the market trading order. Like the Fat Finger incident, illegal behaviors, lack of processes and loopholes in trading mechanism are the initiators of such incidents.