An unfair trading behavior is to use the advantages of network technology to "seize" the transaction before the relevant information is spread to the market. Such a transaction is not only unfair, but also suspected of insider trading. For example, "news aggregation" is suspected of insider trading. The algorithm searches for keywords in the Internet, news websites and social media, and throws a large number of orders in a few milliseconds. The transaction speed is fast, and it is generally completed before the relevant information is widely disseminated. Therefore, foreign regulatory authorities are discussing whether such transactions violate insider trading rules.
Another improper trading behavior is to drive up the stock price, that is, through repeated declaration and withdrawal, to induce other investors to continuously raise the declared price, so as to quickly push up the stock price in a short time and profit from it. 20 10 the Dodd-frank act of the United States explicitly prohibits fraudulent transactions. 2015165438+10.3, the jury of the federal court in Chicago, USA, ruled that MichaelCoscia, a high-frequency trader, was found guilty of fraud and fraud in commodity trading. This is the first criminal case in the United States that uses the anti-fraud transaction law to determine the crime of fraud. In the design process of quantitative trading system, one or more independent quantitative trading strategies should be avoided to frequently declare or cancel orders for specific trading products for the purpose of closing positions, so as to prevent fraud.
The third kind of improper trading behavior is self-trading. By controlling the self-trading behavior of buying and selling in different accounts, it will send a wrong signal to the market, resulting in the market price and liquidity can not accurately reflect the real level of the market. The Commodity Futures Commission of the United States analyzed the trading data from 20 14 to 20 15, and found that the self-trading behaviors between companies, including on-site trading and off-site trading, accounted for a large proportion of the intraday trading data of a single futures contract. For example, in February of 20 15, the self-trading behavior among companies in a futures contract accounted for 10% of the total trading volume of the contract. If calculated by days, this self-trading behavior accounted for 15%. Self-dealing in other contracts only accounts for 5%.
The prevention of the above misconduct should be limited by setting up internal procedures of the trading system, so as to protect quantitative traders from being punished by the regulatory authorities for compliance risks when running multiple quantitative trading systems at the same time.