I. Definitions
Algorithmic trading refers to the use of computer programs and mathematical models to determine the timing, price and quantity of trading orders according to the historical data analysis of securities, real-time market conditions and strategies and parameters selected by traders, etc., by splitting large orders into small orders, in order to reduce the cost of market impact, improve transaction efficiency and transaction concealment of intelligent transaction execution methods, is a perfect combination of manual trading and computer-aided trading system.
Second, the development of intelligent algorithms
The initial algorithmic trading, just to follow the average market price, slowly developed into the passive transaction through pending orders to pursue a better average trading price, with the introduction of artificial intelligence, machine learning and other technologies, gradually evolved into some combined with intelligent signals of active algorithms, completely rely on intelligent signals of intelligent algorithms. The requirements for the speed of market and transactions are getting higher and higher, and the requirements for machine performance are getting higher and higher.
The intelligent algorithm center covers a variety of trading scenarios for customers, provides a variety of algorithm choices, and continues to introduce market-performing algorithms to build one-stop algorithm services for customers.
3. Common intelligent algorithm strategies and their application scenarios
1. Volume Weighted Average Price Strategy - VolumeWeightedAveragePriceVWAP
It is to refer to the historical volume distribution of the security and combine the real-time market order opening algorithm within the specified time range, aiming to make the average transaction price in the parent order trading session as close as possible to the average price weighted by the market in the corresponding time period.
Applicable scenarios: Securities with regular transaction distribution such as large-cap stocks are often used for smooth completion within a specified period of time, position adjustment transactions, block reductions, share repurchases, etc.
2. Trading Time Weighted Average Price Strategy - TimeWeightedAveragePriceTWAP
It is an algorithm that evenly splits orders by time within a specified time range, aiming to make the average transaction price in the parent order trading session as close as possible to the arithmetic average price of the market in the corresponding time period.
Applicable scenarios: inactive and illiquid securities, such as ChiNext; Basket trading, evenly completed within a specified time, position adjustment transactions, arbitrage transactions, etc.
3. VolumeParticipation strategy
The following volume is an algorithm that participates in market transactions according to a certain proportion set by the user, that is, the ratio of the trading volume of the parent order to the total market transaction volume in the corresponding time from the running time is close to the proportion set by the user. Follow-up is a market-driven strategy type. When the market is released, the trading volume will be increased accordingly; When the market shrinks, the trading volume will be reduced accordingly, and the market transaction will be participated in strictly according to a certain proportion of the market volume.
Applicable scenarios: Quickly open or adjust positions according to a certain market proportion, control the volume ratio, and avoid large market impacts, large reductions, share repurchases, etc.
4. Call strategy - PINLINE
The price following strategy is a market-driven strategy that participates in market transactions according to a certain percentage set and intelligently optimizes prices. Relative to the reference price (default is the market VWAP in the past period), when the market price is favorable, increase the volume ratio; When the market price is unfavorable, the volume ratio is reduced accordingly.
Applicable scenarios: Scenarios that participate in market transactions according to a certain proportion of market trading volume and need to optimize execution prices.
5. Quick Strategy - DMA
Express policies are proactive. Designed to complete trade execution as quickly as possible, taking into account market shocks and regulatory requirements.
Applicable scenarios: Scenarios where the amount of individual stocks or baskets is not very large is quickly completed.
6. ICEBERG - ICEBERG
Iceberg is a functional strategy, hanging a certain percentage of the amount on the set price, the pending order has a transaction or the market price changes and then constantly replenishes the order, and maintains a certain transaction progress according to the length of the transaction time, so as to be able to complete the order, the main advantage is that the trading intention is not exposed during most trading hours.
Applicable scenarios: suitable for orders with a large amount, complete in the form of passive orders at the preset target price as much as possible to avoid exposing the real trading volume.
7. Swap
Swapping is a functional strategy, which can realize the operation of buying and selling at the same time when the existing funds are insufficient, and this strategy is only available in one-click trading. Swapping is based on the average price strategy to increase the unified regulation of the transaction amount of the transaction.
It addresses the needs of the following scenarios:
A. Buy only the funds sold and concluded, that is, the original available funds without using the fund account (regardless of commissions and other fees)
B. Lock in the upper limit of the accumulated bid-ask deviation (i.e. the difference in transactions).
Applicable scenario: There is a demand for trading at the same time.
Fourth, the diversity of intelligent algorithm services