Alpha Library: A high-performance rolling window calculation library implemented in Rust with Python bindings. Used for financial data analysis and factor research.
-
Updated
Sep 7, 2026 - Rust
Alpha Library: A high-performance rolling window calculation library implemented in Rust with Python bindings. Used for financial data analysis and factor research.
Auditable multi-agent A-share factor research, knowledge base, long-only backtesting, AutoCombine and QuantCombine.
众人的因子回测框架 stock factor test
An Agentic AI Framework for Empirical Research in Quantitative Finance
Open-source Japanese equity quant research stack for J-Quants/local data, factor & ML research, backtesting, portfolio/risk, paper simulation and MCP agents.
Polars-native, Rust-accelerated factor research engine with 450 built-in alphas, leakage-aware evaluation, and interactive reports.
Research-reproduction Agent: PDF → factor code → backtest → Red Team → reproducibility score. Part of the alpha-kit stack.
Polars-first factor research and backtesting toolkit for quantitative finance.
面向中国 A 股研究的 Agent Skill:公告、财报、研报解析与因子候选生成
Agent-native quantitative research harness. Help agents develop and improve research plugins, run experiments, and analyze traceable results through CLI, MCP, and Studio.
AlphaLattice: Local multi-agent workstation for quantitative research. Your Claude Code or Codex agent directs specialists (Data, Factor, Alpha, Risk, Portfolio, CRO) from factor discovery to a CRO-reviewed portfolio, with reproducible evidence.
Local-first Rust toolkit for quantitative expression search: typed genetic programming, semantic deduplication, feedback-guided search, and causal factor evaluation. Reproducible artifacts, no cloud required.
面向 Crypto 市场研究的 Agent Skill:现货、合约、链上与协议证据解析
Enterprise-grade quantitative factor research terminal - Atelier-grade design - Numba DSL - DES backtester - 87% test coverage
Backtest overfitting audit for factor research: probability of backtest overfitting (PBO), deflated Sharpe ratio, point-in-time data, purged walk-forward. Searches published factor libraries and reports what actually survived costs.
Evidence-linked factor research workbench with bounded model proposals, persistent workers and reproducible synthetic evaluation.
Chill quant research for people who don't speak Python. Ask in Chinese, get backtests.
Python/Parquet/DuckDB tooling for reproducible KuCoin futures data and point-in-time research universes.
Compact Python sample for leakage-aware multi-asset factor construction and evaluation.
A 股量化研究的最小可信内核:先修数据地基,再谈策略结论(MIT)
To associate your repository with the factor-research topic, visit your repo's landing page and select "manage topics."