📈This repo contains detailed notes and multiple projects implemented in Python related to AI and Finance. Follow the blog here: https://purvasingh.medium.com

Advanced Quantitative Factor Research: ML-powered stock return prediction with 72% performance improvement. Features comprehensive alpha factor library, systematic feature selection, and deep learning models (LSTM+ResNet achieving IC=0.06476).

Build a statistical risk model using PCA. Optimize the portfolio using the risk model and factors using multiple optimization formulations.

A deep reinforcement learning framework for generating formulaic alpha factors for quantitative investment, powered by GFlowNet, implemented in Python&PyTorch.

Contains detailed and extensive notes on quantitative trading, leveraging NLP for finance, backtesting, alpha factor research, portfolio management and optimization.

Machine Learning Enhanced Multi-Factor Quantitative Trading: A Cross-Sectional Portfolio Optimization Approach with Bias Correction

Rep tho share codes and projects from the Artificial Intelligence for Trading Algorithms course @udacity.

📈This repo describes a framework that leverages sentiment stability of a financial 10-K report as the trading signal (alpha factor)

📈This repo describes a framework that leverages sentiment stability of a financial 10-K report as the trading signal (alpha factor)

Advanced GAS Price Level Indicator combines machine learning-inspired alpha factors with traditional technical analysis, providing adaptive, correlation-weighted signals for high-probability trading opportunities.

Multi-factor quantitative equity trading system with mean-variance optimization and market regime detection

Construction of PCA class from scratch and 3 implementations of PCA.

End-to-end quantitative trading system for Taiwan stocks — DoubleEnsemble model, 300+ factors, Walk-Forward backtesting, React dashboard

RR-Agent — 自研 A股量化研究工作台 · 因子库 · ML 选股 · CPCV/DSR 回测 · 组合优化 · 算法执行。Self-developed quantitative research workbench for China A-shares: in-house factor library, ML stock selection, CPCV+DSR-validated backtesting, industry-neutral portfolio optimization. Built on multi-source-validated ReachRich data. 不构成投资建议。

LLM-Seeded Evolutionary Discovery of Robust Quantitative Trading Signals (NeurIPS 2026)

Production-grade pipeline transforming raw OHLCV into 77 stationary, ML-ready features. Includes anti-leakage validation, TimeSeriesSplit selection, and 6 signal families for financial time series.