Graph-based stock correlation and prediction for high-frequency trading systems

High-frequency trading (HFT) can be used to achieve profitable performance. It is used to execute a significant number of orders within a short period. Advanced algorithms are created to earn profits with low risks.

One of the approaches used in HFT is based on stock correlation analysis which is used in a machine learning model named Graph Attention Long Short-Term Memory to form price prediction.

The name of the model includes the main principles it is based on.

The first one is the multi-Hawkes Process, which is used to build a correlation graph between stocks. The statistically significant correlation between the stocks is selected then.

Then the weighting matrix underlying the dynamic graph is formed. The price prediction is made then.

The above-mentioned steps are incorporated into a portfolio management model that enables to evaluate and select the proper risk-return combination of assets. 2 portfolios are constructed: a long position in shares and a short position in stock index futures. Such a model can give the result of an annual return of 44.71% with a daily risk of only 0.42%.

High-frequency trading based on advanced machine learning (neural networks in this case) is a sophisticated but highly-rewarding trading method. It needs high expertise and proficiency. The team of ISEC Wealth Management is proud of the experts and services it offers. Implementing state-of-the-art techniques and tools is the core value of the company.


Risk Warning: The information in this article is presented for general information and shall be treated as a marketing communication only.  This analysis is not a recommendation to sell or buy any instrument.  Investing in financial instruments involves a high degree of risk and may not be suitable for all investors. Trading in financial instruments can result in both an increase and a decrease in capital. Please refer to our Risk Disclosure available on our web site for further information.

Related Posts

2 thoughts on “Graph-based stock correlation and prediction for high-frequency trading systems

  1. Worth to invest with ISEC WM. It is one of the most transparent wealth management companies I have ever seen in my entire life and that is very respectful because sometimes they are just thinking about their own wealth but not about their clients’ gain! And they do not ask for a lot of commissions as others do. That is why they are my choice!

  2. Wow, this article explains how high-frequency trading works in a very easy words, at least for me. Long story short, it’s all about using super smart computer programs to make lots of quick trades and predict stock prices. They even use a special kind of math to see which stocks are connected, which is impressive. Then they use this info to make smart guesses about prices. I split them into two groups: one they think will go up and the other they think will go down! I am proud of ISEC Wealth Management and the experts who can do it so well!

Leave a Reply

Your email address will not be published. Required fields are marked *

RSS
Follow by Email
LinkedIn
Share
Instagram