About

Project Introduction

This project implements a production-level investment strategy using a hybrid model that combines VECM (Vector Error Correction Model) and EGARCH (Exponential GARCH). Beyond simple backtesting, it simulates real trading environments and achieves more accurate predictions and risk management through confidence-based position sizing and dynamic reoptimization strategies.

Project Goals

Bridging Theory and Practice: Transforming time series analysis theory into actual trading strategies

Open Source Contribution: Publishing all algorithms as open source to grow together with the community

Educational Content: Providing detailed explanations of theoretical background and implementation processes through Udemy courses

Transparent Performance Disclosure: Publishing actual trading history and performance metrics to demonstrate strategy reliability

Project Structure

Section	Content
Section 1	Financial Time Series Analysis - From fundamentals to advanced models in time series analysis
Section 2	Advanced Investment Strategy Design - Practical investment strategy design and implementation
Section 3	Production Investment Strategy - Production-level trading simulation
Section 4	Advanced Time Series Models - State-space, Kalman, Prophet, LSTM, tree-based ML, wavelet, and copula strategies with walk-forward engineering
Section 5	Factor-Based Asset Pricing Models - From CAPM limitations to Fama-French multi-factor models and practical portfolio construction

Resources

GitHub Repository

View the complete source code on GitHub. Open source and community contributions are welcome.

Visit GitHub

Udemy Course

Learn the theoretical background and detailed implementation process step by step. A comprehensive course including practical examples and advanced techniques.

Enroll Now