Semi-automated trading is a system that uses computers combined with expert input to optimize the decision-making process. Traders are primarily responsible for the decision-making process. They process information and data that computers haven’t solved on their own, while computers bring speed, stability, and accuracy. There are many semi-automated trading systems to support investors. Let’s take a look at examples in Vietnam’s stock market from a simple to complex level.
Stock filter – Complexity: very simple. It is a widely used application of semi-automated algorithmic trading. The trader defines a set of criteria and the system will find qualified stocks. This saves a lot of time by shortlisting the market-wide search to several potential stocks.
In Vietnam, there are misconceptions about equating stock filters and algorithmic trading. Stock filters only help to the degree of data query and give supporting information. In terms of complexity, a professional investor with strong Excel skills can build a good filter in 2 days. However, building a complete algorithmic trading system requires several years along with a solid understanding of database management, server administration, programming (Python, C), investment strategies, and risk and portfolio management.
Open/close Position Schedule – Complexity: simple. The investors clearly understand the market situation and what actions to take if certain market conditions are met. Constantly checking market data to wait for conditions can be time-consuming. This leads to the use of a semi-automated trading system. Some common actions in this setup are price momentum trading, stop-loss, or take-profit.
Trading ecommendation – Complexity: average. While not common in Vietnam, this approach allows systems to suggest trading ideas. It requires an investor’s approval to execute orders. It always leaves the final decision up to the traders.
Parameter Configuration – Complexity: complex. It is an algorithmic trading system where investors can interact with the system in real-time by changing algorithmic parameters. They can thus impact trading performance. The system may be fully automated using a default configuration. This setup is highly effective when expert experience is extremely valuable, or when the system is in risky situations that require human intervention.
Multi-Asset Class rading – Complexity: very complex. Even when investors have great trading ideas, like pair trading for example, they will have difficulties realizing the ideas without the support of an automated trading system. Thus, a predefined set of rules that allows parameter configuration will be extremely useful for investors who want to implement a pair trading strategy.