factor-model

Package to build risk model for factor pricing model

Data Science Project: Replication of "Forest Through the Trees: Building Cross-Sections of Stock Returns" - creation of assets to test validity of factor models with Python

This repository shows the application of PCA technique for risk factor modelling of financial securities.

璇玑 XUANJI · AI 自主进化 A 股量化模拟盘系统 — 覆盖数据采集→因子计算→策略回测→模拟执行→风控监控→AI自主调度/复盘完整闭环。五层AI Agent架构，硬风控+白名单执行，可审计可回滚。

八层风控 · 37因子自进化 · A股全自动量化交易系统 | 16 modules · 6400+ lines · AI-driven

Repository for the AugmentedPCA Python package.

Package to build universes for factor pricing model

Barra-style multi-factor risk model & risk attribution: cross-sectional WLS factor returns, Ledoit-Wolf shrinkage covariance, factor vs specific risk decomposition. Offline via a panda_data adapter. Research/education only, not investment advice.

A multi-factor equity scoring framework that goes beyond cheap vs. expensive. Models expectations, quality, reflexivity, crowding, and regime fit as orthogonal factors.

Multi-layer investment agent framework for A-share equities | A股多层投资智能体框架 — Fisher → Sharpe-Fama-Merton → Graham → Markowitz + Damodaran

A股量化选股系统 V2 — 三阶层选股 + WxPusher 微信推送

This is a tentative pytorch implementation of the paper "Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders"

Portfolio research on US equities — point-in-time data, survivorship-bias-free backtests, walk-forward validation gated by Deflated Sharpe and PBO. 158 factors over 20,931 tickers (1997-2026), plus tactical ETF allocation. Ships the rejections too: 1 adopted, 20+ rejected, and one headline number retracted.

Sector-relative equity valuation model using Ridge-weighted factor scores and within-sector z-scoring across 11 GICS sectors.

Replication code for "The Shape of Beta: Industry Factor Structure and Crisis Risk Premium" (Woo & Kim, 2026)

Multi-factor risk model (Momentum, Size, Volatility, Value, Beta) built on cross-sectional regressions. Estimates factor covariance, specific risk, Euler risk contributions, and active tilts. Interactive Streamlit app included.

A PyTorch research pipeline for cross-sectional stock return forecasting with grouped factors and stock-wise attention.

量化交易 Adaption Kitset，BoBanana 5.0 配套的量价因子 + RL 工具集

此為機器學習與財務計量專案的環節之一