Why does Algorithmic Trading exist?

Algorithmic Trading is a trading method that uses computer software that has been programmed with trading methods or algorithms at the writer's discretion. This trading method can theoretically achieve profits at a speed and frequency that humans cannot keep up with. Algorithmic trading is an important element that constitutes the concept of High-Frequency Trading (HFT) or High-Frequency Trading.

The computer software used in algorithmic trading is often programmed with trading rules based on time, price, quantity or any other mathematical model. In addition to generating profits for the writer (or investor), algorithmic trading can increase market liquidity and eliminate the emotional impact of the trader in each order.

Algorithmic trading example

Suppose an investor wants to apply the following trading criteria:

1. Buy 50 stocks with code ABC on the stock exchange when its 50-day moving average is above the 200-day moving average. (A moving average is an average of past price data points that helps daily price fluctuations and thereby identify trends)
2. Sell stock ABC when its 50-day moving average is below its 200-day moving average. With the above two conditions, a computer software will automatically monitor stock prices (calculate moving averages) and place buy and sell orders when the conditions are satisfied. Investors no longer need to monitor prices and charts to time orders manually. Algorithmic trading software helps significantly reduce the effort investors have to spend to focus on other things.

How is algorithmic trading programmed?

To develop an algorithmic trading software capable of generating profits for investors, the programmer must do the following steps:

Strategic identification

This is the most important step in the entire development process. Investors must research the trading strategy (for example, what conditions must be met at the entry point), and especially risk management conditions to avoid losses exceeding the allowable limit in case of loss. in case the market changes unexpectedly.

Programming Software

After identifying the strategy, the programmer will rely on the conditions or trading instructions to create an algorithmic trading software.

Algorithmic trading software can be written in many programming languages such as Python or Go. After that, these software will be installed on the cloud server or run directly on personal computers. The characteristic of installing software on a cloud server is that the software can run 24/7, instead of having to turn on the computer regularly like a personal computer.

In addition, the software must be able to interact with the exchange or stock brokerage company through API (Application Programming Interface) or supported protocols. Another thing to note is the latency of information exchange between the software and the server of the exchange or stock brokerage company. There are many options that can help reduce latency but will be covered in a later article on High-frequency trading.

Testing (testing)

After algorithmic trading software has been written, investors must test the software in two ways:

Experiment with a simulated environment (paper trading): Trading software can be tested on a simulated trading environment (or simply paper trading) with the latest market data transmitted. If the delay of data transmission is minimal, this is a very effective way of testing trading software.

Testing with past data (backtesting): Trading software can also be tested using historical price data (also known as backtesting) to determine whether the software is working as expected, or whether there is potential to generate profits. profit if the market performs as predicted.

However, it should be noted that running tests on past data without careful care can result in the trading software over-optimizing for market outliers. Therefore, the testing plan must include many stages, a lot of data, and be closely monitored by investors.

Implement real transactions

If the trading software has completed the above steps, investors can enter the final stage of starting real trading.