Modeling High-Frequency Limit Order Book Dynamics Using Machine Learning

Providing the solutions for high-frequency trading (HFT) strategies using data science approaches (Machine Learning) on Full Orderbook Tick Data.

  • Framework to capture the dynamics of high-frequency limit order books.

pipline

Overview

In this project I used machine learning methods to capture the high-frequency limit order book dynamics and simple trading strategy to get the P&L outcomes.

  • Feature Extractor

    • Rise Ratio

Price_B1A1

  • Depth Ratio

depth

[Note] : [Feature_Selection] (Feature_Selection)

  • Learning Model Trainer

    • RandomForestClassifier
    • ExtraTreesClassifier
    • AdaBoostClassifier
    • GradientBoostingClassifier
    • SVM

  • Use best model to predict next 10 seconds

CV_Best_Model

  • Prediction outcome

prediction-1

  • Profit & Loss
    P_L

    [Note] : [Model_Selection] (Model_Selection)

GitHub

https://github.com/rorysroes/SGX-Full-OrderBook-Tick-Data-Trading-Strategy