github

Having a background in mathematical finance, I have always been extremely passionate about analyzing the market movements and the stock trends. This is an ongoing project that I have been doing for several years, with many ideas implemented. I also treat this project as an opportunity to practice with many different Deep Learning models. In this project, I have achieved:

• Collect and perform proper data engineering on stock market prices of the S&P 500.
• Create and build strategies to maximize trading expected return using black box Bayesian optimization, Convolutional Neural Networks (CNN) and Deep Q-Learning using Tensorflow, Pytorch and OpenAI Gym; invent new neural networks based on transformer structure for noisy financial input; design suitable environments and memory controls for the models.

The implementation of my ideas is shown in the github linked above. Even though the documentation is not in details, you can find the up-to-date samples of my coding there. I plan to keep working on this repository as an ML models practice.