pandas is an open source, BSD-licensed library providing high-performance, easy-to-use data structures and data analysis tools for the Python programming language.
Contingency Random Number Generator that produces random numbers with real financial market statistical signatures (fat tails, volatility clustering, kurtosis). Matches 86% of real market metrics vs 14% for NumPy.
eXtensible Time Series: Provide for uniform handling of R's different time-based data classes by extending zoo, maximizing native format information preservation and allowing for user level customization and extension, while simplifying cross-class interoperability.
Extension of data.frame: Fast aggregation of large data (e.g. 100GB in RAM), fast ordered joins, fast add/modify/delete of columns by group using no copies at all, list columns and a fast file reader (fread). Offers a natural and flexible syntax, for faster development.
Black-Scholes and CRR-binomial options pricing with Greek-attributed backtest P&L, decomposing each trade into delta, gamma, theta, vega, vanna, and volga via a second-order Taylor expansion.
A Python Finance Library that focuses on the pricing and risk-management of Financial Derivatives, including fixed-income, equity, FX and credit derivatives.
Stochastic path simulation (Heston, rough Bergomi, Kou, curve-fitted Hull-White and G2++, CIR, commodity and credit models) with Monte Carlo pricing, exposure/CVA, risk-factor backtesting and a model-validation lab producing interactive Plotly HTML reports.
Vanilla and exotic option pricing library to support quantitative R&D. Focus on pricing interesting/useful models and contracts (including and beyond Black-Scholes), as well as calibration of financial models to market data.
Toolkit for option pricing, implied volatility surfaces, risk-neutral densities, and pricing kernel surfaces with support for advanced models including Heston, Kou, and Bates.
High-performance pricing engine built on QuantLib. It exposes QuantLib's functionality through gRPC and REST APIs, enabling distributed computations with FlatBuffers serialization.
Free quant finance API with 63 deterministic endpoints + 15 free interactive calculators at [quantoracle.dev](https://quantoracle.dev). Options pricing with full Greeks, Monte Carlo, Kelly, VaR, Sharpe, CAGR, crypto liquidation, impermanent loss, plus live crypto volatility/funding data and 24/7 position monitoring with webhook alerts. 1,000 free calls/day, no API key.
Multi-factor quantitative stock analysis MCP server for US, HK, and CN A-share markets. Transparent 0-100 scoring from 40+ indicators. Listed on Official MCP Registry.
Black-Scholes-Merton pricing, Monte Carlo with antithetic and control variates and confidence intervals, path-dependent payoffs and finite-difference Greeks.
SOFR compounding-in-arrears per ARRC/ISDA conventions (lookback, observation shift, lockout) and the SOFR Index method; reproduces the NY Fed's published averages.
Quantitative finance library with 130+ stochastic processes, option pricing and calibration, volatility surfaces and copulas, SIMD/GPU accelerated with Python bindings.
A Python module for fitting the [Log-Periodic Power Law Singularity (LPPLS)](https://en.wikipedia.org/wiki/Didier_Sornette#The_JLS_and_LPPLS_models) model.
Polars-native technical analysis and backtesting with bit-identical batch and streaming parity, plus an agent skill for consistent research-to-live strategy code.
Financial market primitives in Rust: Price/Quantity/Symbol newtypes, BTreeMap order book, OHLCV aggregation, SMA/EMA/RSI indicators, position ledger with PnL, and composable risk monitor.
Stateless OHLC analyzer: POST bars and requested indicators, get back RSI/MACD/Bollinger/ADX/ATR/VWAP/Ichimoku plus smart-money-concept primitives (order blocks, FVGs, BOS/CHoCH, swing structure). No database, no AI signals.
Daily dealer-gamma levels (gamma flip, call and put walls, max pain, expected move) for DAX, Nasdaq, Dow and gold, computed from options open interest (Eurex ODAX for the DAX, ETF proxies for the US) and written into a TradingView Pine v6 indicator; the method is documented in open studies on the author's site. [Website](https://kruegeralgorithms.com/en/research).
Self-hosted platform for AI-assisted Python strategy development, backtesting, and paper and live trading across crypto exchanges and traditional brokers.
Local-first macOS workbench for A-share and US stock strategy screening, paper trading, position sizing, and risk alerts with read-only broker integrations.
Trading-strategy research framework with iterative proposal generation, backtesting, SQLite memory, holdout evaluation, walk-forward experiments, and experimental genetic-algorithm and differential-evolution optimizers.
Catalogue of twenty-seven position-exit policies (stop-loss, take-profit, time, volatility, signal-reversal and convergence) behind one interface, with a drop-in adapter for backtesting.py.
Estimates fees and impermanent loss for concentrated-liquidity pools whose volatile assets can be hedged, then backtests whether rotating capital among pools outperforms remaining in one pool. It does not execute trades or manage liquidity.
Embeddable limit order book and matching engine with integer-exact pricing, a single-writer core and write-ahead-log crash recovery, plus a microstructure research harness whose order-flow-imbalance, Kyle's lambda and CVD studies are measured against simulator ground truth.
Streaming backtesting engine for short-duration index options with conservative five-minute execution modeling and reproducible Early Retirement Now and WealthyOption strategy replication results.
Multi-agent paper-trading framework where LLM agents author orders and a separate broker process enforces fifteen fill-time safety rails; each fill is stamped with the git commit it executed against for reproducibility.
Self-hosted market-alert daemon: price, RSI/MACD/SMA conditions, and full Pine Script v5 custom indicators evaluated off-TradingView, on crypto (Coinbase) and stocks (Yahoo Finance) with no API key, delivered via ntfy push, email, or Telegram.
Robinhood API client for unattended automation: after the first approved login, sessions renew from a stored refresh token with no password or device approval prompt. Covers stocks, equity and index options with Greeks, futures, IRA accounts, and the official Crypto Trading API.
Scores detected chart patterns against the pattern-free baseline for the same market, timeframe and horizon, reporting lift with cluster-robust confidence intervals instead of a hit rate against 50%.
Linter for mechanical trading-rule conditions: replays every condition over historical bars to catch look-ahead levels, dead branches that can never fire, and regime-drifted absolute thresholds before you trust a backtest.
Cross-asset research and paper-trading engine with point-in-time data, walk-forward validation, multiple-testing accounting and a published forward record.
Conversational multi-agent quant framework where agents rank currently-effective factors for entry timing (time-series rank IC), write complete strategy code that passes sandboxed audit before backtesting, and evolve strategies under multi-objective fitness; every order requires machine approval and the LLM never has a direct order path.
AI auto-trading bridge for East Money's miaoxiang (妙想) China A-share simulation platform; BYOK multi-LLM (OpenAI/DeepSeek/Moonshot/GLM/Qwen) decision brain → automated order placement via miaoxiang API, with daily cron review and weekly AI reflection.
Multi-agent LLM trading analysis where 12 AI agents (analysts, debaters, risk manager) debate stock picks in real-time, supporting US equities and China A-shares.
Autonomous trading system for crypto linear perpetuals in which language models only propose and deterministic rules decide: code computes every figure, prices the full round trip before an order exists, sizes from the operator's risk at the stop, keeps exits at the exchange, and records each cycle with a per-file SHA-256 that can be re-verified.
AI trading operating system with 9 ML models (RF, XGBoost, EMAformer, MOIRAI-2, Chronos-2, DDPM, PPO RL, GNN, LR) ranked by Sharpe ratio, Claude AI synthesis with dual-tier routing (~$0.003/analysis), real-time Next.js dashboard, Alpaca paper trading, and 277 automated tests.
Natural-language multi-agent finance research agent with 29 swarm presets, 70 skills, and 28 auto-discovered tools; 7 backtest engines covering A-shares/US/Crypto/Futures/Forex/Options plus a cross-market CompositeEngine with shared capital pool; 5-source auto-fallback data layer (tushare/okx/yfinance/akshare/ccxt); 17-tool MCP server; includes trade-journal behavioral diagnostics for 同花顺/东财/富途 exports.
MIT-licensed desktop terminal for OKX USDT perpetuals that pairs live discretionary trading with systematic research: realtime WebSocket market data, 24 indicators, a market-wide radar, AI assistants with typed wake conditions, and a bundled Python runtime for 1-minute backtests. Permission modes (advisor / copilot / limited_auto) are enforced by the runtime rather than by the prompt, and demo and live credentials are isolated.
Self-hosted execution engine for algorithmic trading bots. Write strategies in Python, TypeScript, Rust, C++, C#, Scala, or Haskell and deploy with one command. Each bot runs in an isolated container with scheduled or streaming execution.
Tamper-evident SHA-256 hash-chain audit log and human-readable report for AI trading-agent decisions; read-only, offline, and dependency-free (Python standard library only). [GitHub](https://github.com/Autonomous-Asset-Management-Agents/autonomous_/tree/main/packages/autonomous-audit).
Algorithmic trading framework where the same code runs for backtesting and live trading across stocks, options, crypto, futures, and forex with multiple brokers including Alpaca, Interactive Brokers, Tradier, and Schwab.
macOS desktop workspace where your own Claude Code or Codex writes a trading strategy, backtests it with a Monte Carlo permutation test, a parameter-plateau scan and walk-forward, and runs it live through deterministic scheduled code with no LLM in the order loop (Binance, OKX, BingX, Gate.io, Bybit, paper trading).
Actively maintained, high-performance backtesting and live trading framework with AI-assisted strategy tooling (MCP server, skills, agent, web platform). Fork of backtrader.
Closed-form expected returns, Sharpe ratios, and skewness of trend-following systems, with complete implementations and multi-decade futures backtests.
Checks how much of an LLM trading backtest falls on or before the model's training cutoff and how many clean months a claimed Sharpe needs once the number of attempts is counted, with an MCP server and an evidence-adjusted leaderboard of public AI trading competitions.
Hash-chained pre-registration and falsification ledger for research claims: write down what evidence would kill your claim before seeing the data; append-only JSONL with a sha256 chain (`fl verify` detects any edit), Wilson-CI hit-rate vs the random baseline, fail-closed falsification report contracts. [PyPI](https://pypi.org/project/falsification-ledger/).
the project using sql, pandas to provide an uniform and extendable way to record data, computing factors, select securities, backtesting, realtime trading and it could show all of them in clearly charts in realtime.
AlphaEvolve-style search in which an LLM invents and evolves trading strategies as Python code, judged on sealed holdouts and a pre-registered multi-market exam whose bar corrects for every attempt.
An AI-oriented Quantitative Investment Platform by Microsoft. Full ML pipeline of data processing, model training, back-testing; and covers the entire chain of quantitative investment: alpha seeking, risk modeling, portfolio optimization, and order execution.
Backtester that raises on look-ahead (decide at close, fill at next open), replays funding and corporate actions point-in-time, and reports deflated Sharpe against the full trial count.
Pre-trade check and written audit from your own NinjaTrader trade history: sessions in market time, first trades of the day, re-entries after a loss, exit replay on minute bars, and written rules checked against the chart without look-ahead; never places orders.
pysystemtrade is the open source version of Robert Carver's backtesting and trading engine that implements systems according to the framework outlined in his book "Systematic Trading", which is further developed on his blog.
Options trading system: one event-driven engine for backtesting and live trading on Interactive Brokers, with pricing, volatility and risk research tools.
A high-frequency trading and market-making backtesting tool accounts for limit orders, queue positions, and latencies, utilizing full tick data for trades and order books.
Realistic limit-order fill simulator for options credit/debit spreads with post-and-wait limits, stale-quote guards, deterministic same-bar tiebreaks, and a patient-then-cross exit; engine-agnostic and zero runtime dependencies.
Reproducible negative-result benchmark: no model beats a memoryless synthetic binary-options feed (AUC ≈ 0.50), with permutation-null tests and a look-ahead-leak case study.
A base on Python/C++ open source high-performance quant framework for faster analysis and backtesting, contains the complete trading system components for reuse and combination.
Toolkit for quantitative trading analysis. It integrates an advanced market screener, a multi-strategy, multi-asset backtesting engine. Use with built-in GUI or through CLI.
An open‑source, broker‑agnostic Python library that lets you seamlessly deploy strategies from any major backtesting engine to live (or paper) trading with zero code changes and built‑in safety controls.
A lightweight, event-loop-style backtest engine that allows a function-driven imperative style using efficient stateful helper functions and data containers.
AI-powered trading edge discovery platform that validates trading ideas with event studies, statistical tests, and real market data. Web app, MCP server, CLI (`pip install varrd`), and Python SDK.
Local-first backtesting engine with built-in overfitting checks (PBO, deflated Sharpe, bootstrap CI, walk-forward) and a native MCP server for AI agents.
Dependency-free audit toolkit for backtests (no numpy/pandas): detects look-ahead, checks cash/position ledger conservation, flags fee and funding double-counting, and mutation-tests your own suite for blind spots.
Transaction infrastructure for defining instruments, transactions, portfolios and accounts for trading systems and simulation. Provides portfolio support for multi-asset class and multi-currency portfolios. Actively maintained and developed.
Transaction-oriented infrastructure for constructing trading systems and simulation. Provides support for multi-asset class and multi-currency portfolios for backtesting and other financial research.
Trading framework built around CCXT with Zarr-backed OHLCV persistence, contiguity-checked data feeds, isolated-margin position management, and a loop-based backtester sharing strategy code across simulation, paper, and live modes.
Backtest significance checker with published false-positive rates: dependence-robust tests corrected for every variant tried (Romano-Wolf), deflated and probabilistic Sharpe ratios, probability of backtest overfitting, breakeven costs and a one-bar lookahead check, taking vectorbt portfolios, backtesting.py results and TradingView trade exports directly.
C++ 17 based framework/library (with sample applications) for testing options based automated trading ideas using DTN IQ real time data feed and Interactive Brokers (TWS API) for trade execution. Comes with built-in [Option Greeks/IV](https://github.com/rburkholder/trade-frame/tree/master/lib/TFOptions) calculation library.
A base on Python/C++ open source high-performance quant framework for faster analysis and backtesting, contains the complete trading system components for reuse and combination. You can use python or c++ freely.
Hyperliquid perpetuals order-book fill simulator with fee, slippage and owner-limit checks, an execution-cost breakdown, and receipt verification that re-derives result hashes from the inputs.
Open-source self-hosted AI trading signal platform. Generates buy/sell signals using RSI, MACD, EMA, Bollinger Bands for forex, crypto and commodities. Deployable via Docker Compose. ([Demo](https://tradeclaw.win/dashboard)).
Generates simulated stock markets for testing trading strategies and AI agents, with a limit order book per company and an economy underneath, so orders fill against the book and move prices, and any run can be forked or replayed exactly.
MCP server that gives AI assistants screeners, multi-timeframe technical analysis and walk-forward backtests for stocks, crypto, forex and futures across global exchanges.
A reasoning & decision-trace visible Multi-agent LLM trading-research framework where bull/bear analysts debate each ticker and every decision cites the evidence it rests on.
MetaTrader 5 in a Windows VM (Docker + QEMU/KVM) over REST and MCP: market data, order/position/history management for automated trading and bots, the strategy-tester (backtesting) API, and server-side indicators (RSI/MACD/Bollinger/ADX/VWAP/Ichimoku, order blocks, FVGs). Multi-broker, multi-account.
Interactive Brokers over REST and MCP (FastAPI + ib_async over a Linux-native IB Gateway): market data (quotes, historical bars) plus order/position/execution management for automated trading across stocks, options, futures, forex, crypto and CFDs.
Self-hosted Telegram bot for Interactive Brokers that runs next to IB Gateway: position snapshots and P&L tracking, plus option orders from chat commands or webhook signals, each confirmed in Telegram before it is sent, with market-hours, duplicate-signal and position-limit checks on webhook signals.
Library of 501 agentskills.io-format skills giving AI coding agents trading-infrastructure playbooks for order idempotency, look-ahead-bias elimination, kill switches, execution algorithms and point-in-time data, each with a standalone Python reference implementation and its own unittest suite.
MCP server and NinjaTrader 8 AddOn that give an AI coding assistant a build loop for NinjaScript: offline and in-platform compile with structured errors, chart and log reads, headless backtests, optimization and walk-forward; read-only by default.
Uncertainty-aware portfolio optimization: compare covariance estimates, test allocations across risk scenarios, and see the cost of tighter constraints.
Portfolio optimization framework combining covariance shrinkage, bootstrap aggregation, and parametric scenario modeling, with reproducible notebooks and an accompanying SSRN paper.
An agentic framework that mimics the real quant trading pipeline to find alpha: economic hypothesis, in-sample iteration, and out-of-sample validation.
Zero-dependency calculator for DCA order ladders on crypto perpetual futures that gives the average entry, take-profit, break-even and liquidation price (isolated or cross margin) after every order.
Python library for portfolio optimization built on top of scikit-learn. It provides a unified interface and sklearn compatible tools to build, tune and cross-validate portfolio models.
Optimisation analytics for constructing and backtesting optimal multi-asset portfolios: covariance estimation, rolling optimisation, and performance reporting in one pipeline.
Eiten is an open source toolkit by Tradytics that implements various statistical and algorithmic investing strategies such as Eigen Portfolios, Minimum Variance Portfolios, Maximum Sharpe Ratio Portfolios, and Genetic Algorithms based Portfolios.
Implementations regarding "Advances in Financial Machine Learning" by Marcos Lopez de Prado. (Feature Engineering, Financial Data Structures, Meta-Labeling).
Quantitative performance and risk analytics with 150+ metrics, portfolio optimization, Monte Carlo simulation, and attribution; actively maintained successor to [empyrical](https://github.com/quantopian/empyrical)/[pyfolio](https://github.com/quantopian/pyfolio).
Framework-agnostic risk-management toolkit for systematic trading — position sizing, drawdown control, a composable stop engine, correlation limits, and portfolio exposure caps, with adapters for backtesting.py, freqtrade, and vectorbt.
Tested implementations of the probabilistic and deflated Sharpe ratios, minimum track record length, probability of backtest overfitting (CSCV) and a seeded prop-firm challenge simulator, each folder giving the formula and its source paper.
PortfolioAllocation is a JavaScript library designed to help constructing financial portfolios made of several assets: bonds, commodities, cryptocurrencies, currencies, exchange traded funds (ETFs), mutual funds, stocks.
Reference data and worked arithmetic for position sizing across forex, stocks and crypto futures: contract specifications (pip size, contract size, minimum step) for 131 instruments, pip value and 20-pip loss by lot size, losing-streak equity decay and drawdown-recovery tables, plus the risk-to-units formulas with two worked examples.
Open-source algorithm that scores two US-listed companies against each other on 8 fundamentals computed directly from SEC EDGAR XBRL data (10-K/10-Q); live comparison tool at duelstocks.com.
AI skills for quantitative factor research: discover, evaluate, mine, backtest, and monitor factors through any AI coding assistant. Supports A-share, HK, and US markets.
Backtest-overfitting audit for factor research: CSCV probability of backtest overfitting, deflated Sharpe against the declared trial count, White's Reality Check, point-in-time universe membership and disclosure-date alignment. Ships a worked example in which 24 pure-noise series produce a 1.11 Sharpe and the audit says so.
Screens cross-sectional signals for incomplete-cross-section leakage from staggered data arrival and grades per-signal release timing; ships a known-ground-truth demo reproducing its method papers.
The FactorAnalytics package contains fitting and analysis methods for the three main types of factor models used in conjunction with portfolio construction, optimization and risk management, namely fundamental factor models, time series factor models and statistical factor models.
Solutions for enhancing portfolio diversification and replications of seminal papers with R, most of which are discussed in one of the best investment references of the recent decade, Expected Returns: An Investors Guide to Harvesting Market Rewards by Antti Ilmanen.
LLM-powered A-share stock analysis via P2P agent collaboration. Technical analysis (MA60, volume-price patterns, golden eye), deep research reports using proprietary methodology, and market state summaries. Analysis logic stays private via Skill-as-API protocol.
Pre-analyzed financial news via REST API and MCP for AI agents: per-ticker impact and sentiment, a category, and a 1-10 relevance score on every story, plus structured SEC Form 4 insider data. Free tier, no card.
Automatic discovery of non-trivial statistical relationships across 500+ time series from FRED, World Bank, EIA, and NOAA using mutual information screening, lagged MI directional testing, and FDR correction.
High-performance ETF pattern matching via DTW with cosine pre-filtering. 43× DTW and 58× pattern-match speedup over pure Python using pybind11/C++20. Includes Jupyter notebook with full algorithm walkthrough.
Offline financial time-series diagnostics, correlation spectra, change detection, and forecast evaluation with explicit coverage and input-data rights checks.
Deterministic offline PineScript v6 → C++ backtest runtime, validated trade-for-trade against TradingView (245/246 strict, 0 engine bugs). Runs locally via Docker and is drivable by AI agents through a bundled MCP server.
Sourced, dated dataset of prop-trading-firm daily-loss and max-drawdown rules (static and trailing, lock levels, reset times) for 9 firms and 13 programs, with a zero-dependency JS/Python drawdown calculator.
Dated SCHD payment and holdings snapshots with source URLs, data dictionaries, and Python tools for split-aware dividend windows and ETF company-exposure checks. [Website](https://dividendsteps.com/).
Point-in-time market-data lake where every read takes an explicit as-of, corporate actions filter on when they became knowable, and delisted names are kept.
Collects and stores funding-rate snapshots from five decentralized perpetual-futures exchanges, normalizing rates hourly and flagging markets with less than $2 million in open interest. Intended for historical analysis and cross-venue comparisons; it does not trade.
NASDAQ TotalView-ITCH 5.0 day files to Parquet: BBO, trades, order-by-order messages with the resting side already resolved, and MBP depth; downloads and verifies the free emi.nasdaq.com samples.
MCP server (Model Context Protocol) exposing 17 tools for AI agents to perform quantitative analysis: real-time stock quotes, full technical analysis (RSI, MACD, Bollinger, ADX, Stochastic, ATR, OBV + pattern detection with structured verdicts), crypto prices via CoinGecko, risk metrics (VaR, Sharpe, Sortino, Beta, Max Drawdown), correlation matrix, options chains, earnings calendar, sector rotation, and portfolio analysis. Works with Claude Desktop, Cursor, Windsurf. No API keys for core tools. FastMCP 3.2.
Unified package for collecting OHLC prices from Binance, Bybit, Coinbase, OKX, Kraken, KuCoin, and Bitget (spot & futures) into a DataFrame, with flexible timestamp inputs and a wide range of intervals.
Live stock/ETF/crypto data with AI-generated bull/bear cases and price forecasts, proprietary ML options pricing with probability ITM and fair value, and news bias scoring across 5,000+ sources. Available as MCP server or API. Free tier: 50 queries, no signup.
CLI + MCP server for SEC 13F institutional holdings, Form 4 insider trading, and smart money convergence signals (tickers where hedge funds and company insiders are both buying). Free anonymous tier, no signup required.
Python module to get data from various sources (Google Finance, Yahoo Finance, FRED, OECD, Fama/French, World Bank, Eurostat...) into Pandas datastructures such as DataFrame, Panel with a caching mechanism.
Self-hosted financial data API for European and US markets: prices per listing, cross-checked fundamentals from 18 sources, DCF valuation and Fama/French factors in your own PostgreSQL.
Historical chart pattern similarity search API. 24M+ pre-computed embeddings across 15K+ symbols and 10 years of data using pgvector. Returns forward returns, regime analysis, and pattern detection. Also available as MCP server. [Website](https://chartlibrary.io).
Free API and MCP server for Japanese company financials. Normalizes EDINET XBRL across JP-GAAP, IFRS, and US-GAAP for 3,800+ listed companies with 90 metrics, screening, and securities report text.
Standardized SEC EDGAR financials (income statement, balance sheet, cash flow, 40+ ratios) for ~19,000 US public companies, normalized from XBRL. REST API + MCP server for Claude/Cursor. Free tier.
Money-market and dollar-funding research with public-source collectors, historical observations, separate release and retrieval clocks, and explicit coverage gaps.
Access Japanese timely disclosures (TDNet) via MCP. Retrieve earnings, dividends, forecasts, buybacks, and other filings for 4,000+ listed companies. No API key required.
Open dataset of U.S. public-company equity stakes in Anthropic and OpenAI from primary 10-K / 10-Q / 8-K filings, court records, and press releases. Each row tagged with a confidence flag (V verified, P probable, S speculative).
Self-hosted China A-share data lake that keeps 55 datasets (prices, financials, capital flows, corporate events, futures and options) in local Parquet with incremental updates, point-in-time queries, delisted-stock retention, DuckDB SQL and a read-only MCP server.
Terminal stock dashboard with keyless quotes, candlestick charts and portfolio tracking, plus optional news and SEC filing views powered by the AlphAI API.
Python package for Swiss financial data (SNB Policy Rate, SARON, CHF FX rates, CPI, SMI equities, Confederation bond yields) from official SNB sources.
yfinance-shaped market data for Iranian markets: TSETMC equities, ETFs, indices and option chains, Codal financial statements, TGJU FX and gold, SCI CPI and Iranian crypto exchanges.
Create meaningful OHLCV datasets from knowledge of [exchange-calendars](https://github.com/gerrymanoim/exchange_calendars) (works out-the-box with data from Yahoo Finance).
simple, hassle-free access to price information of financial assets (currently based on yfinance and pycoingecko), including search and a symbol class.
Python package that implements SDMX 2.1 (ISO 17369:2013), a format for exchange of statistical data and metadata used by national statistical agencies, central banks, and international organisations.
finagg is a Python package that provides implementations of popular and free financial APIs, tools for aggregating historical data from those APIs into SQL databases, and tools for transforming aggregated data into features useful for analysis and AI/ML.
Toolkit with 200+ financial metrics including 80+ financial ratios, 30+ technical indicators, 20+ risk and performance metrics and 50+ macro indicators which pulls from Financial Modeling Prep, Yahoo Finance, OECD, GMBD and more.
MCP server and Python SDK for a SEC EDGAR REST API covering XBRL fundamentals, Form 4 insider trades, 8-K events, 13F holdings and filings, where every record carries the source sec.gov filing URL it came from.
Checks that catch nine silent failure modes in SEC filing data, from Form 4 filings that are about a different issuer to 13F-NT counted as a position report, each documented with the incident that found it.
Financial intelligence API with verified market data, trading signals, sentiment analysis, and fact-checking across 1,061+ tickers. [PyPI](https://pypi.org/project/veroq/).
Real-time market data streaming in Rust: lock-free SPSC ring buffer, 100K+ ticks/second ingestion, multi-timeframe OHLCV construction, and Lorentz transforms on financial time series.
API server for crypto and market data with technical indicators, on-chain analytics, and MCP access; includes x402 payment support for hosted endpoints.
NinjaTrader 8 .ncd tick and minute history files to Parquet without NinjaTrader running; row-exact against NinjaTrader's own export, incremental re-runs.
Trading engine and MCP server for prediction markets with fee-aware no-arbitrage checks across related event contracts, statistical arbitrage, and live or paper execution on Kalshi, Polymarket, and Solana.
Autonomous weather-market trading agent for Kalshi and Polymarket that learns per-station forecast bias from leak-free hindcasts and sizes positions with fractional Kelly under hard risk limits.
Real-time tennis scores, serving and break-point state, and model win probabilities for pricing tennis event markets, plus H2H, rankings and a 1968-2022 point-by-point archive; free tier.
Open-source public record of stock predictions: commit-reveal before the open, automatic settlement against exchange data, tamper-evident hash-chained ledger. Live at [quantrank500.com](https://quantrank500.com).
Tick-level history for Polymarket and Predict.fun crypto Up/Down markets, including the full-precision Chainlink settlement feeds and each market's strike and settled outcome, with a sandboxed runner that replays a submitted strategy against the same archive.
High-frequency trading and market-making backtesting for Polymarket, written in Rust, with queue position, latency and maker/taker fee models applied to tick-level L2 order book and trade data.
Browser charting app for crypto, stocks and forex with indicators, a scripting language with strategy backtests, alerts, bar replay and an MCP server for AI agents.
Technical-analysis charting app for Interactive Brokers (IBKR) that exports drawings, notes, indicators and OHLCV as JSON/XLSX/CSV, and exposes chart context to AI agents over MCP.
Financial charting library for the web, drawn with Canvas 2D or WebGL: 18 chart types, 111 indicators, drawing tools, live exchange feeds, orders on the chart, replay and backtesting.
The ExPy add-in allows easy use of Python directly from within an Microsoft Excel spreadsheet, both to execute arbitrary code and to define new Excel functions.
Agent-native quantitative research harness with plugin-based tasks for data processing, factor analysis, model training, and backtesting, traceable artifacts, and CLI and web interfaces.
Linter and CI for written investment theses that fails a number without a filing citation, freezes test thresholds once results are out, and Brier-scores forecasts.
Multi-agent financial research framework combining fundamental, technical, news, and sentiment analysis with structured investment debates and risk assessment.
Local-first quantitative research and backtesting platform with data synchronization, multi-asset screening, factor analysis, portfolio optimization, risk analysis, and optional LLM-assisted news analysis.
Agent-native quantitative research toolkit for DeepSeek Harness: 46 tools across data, alpha, ML, risk, execution and ecosystem domains, with an end-to-end research pipeline.
Local-first AI research workbench for China A-shares: unified market overview, theme radar, limit-up ladder and sentiment-cycle analysis, AI stock reports, portfolio inspection, and agent-based collection of post-market commentary with evidence tracing.
A dockerized Jupyter quant research environment with preloaded tools for quant analysis, statsmodels, pymc, arch, py_vollib, zipline-reloaded, PyPortfolioOpt, etc.
Local research workbench that a coding agent drives from one sentence: dated US large-cap research positions bounded by each date's information cutoff, SEC-filing evidence review, and an investment committee that changes no numbers.
Local-first, agent-agnostic trading journal and review harness that freezes point-in-time run envelopes and evidence packs, replays them deterministically, and evaluates rule candidates against a frozen 240-case finance benchmark, with no order or broker access.
Open-source transpiler and runtime that executes Pine Script logic in Node.js and the browser with 1:1 syntax compatibility, for running indicators and strategies on your own infrastructure.
Quant interview prep with 391 free problems with full worked solutions, per-firm online-assessment guides, and free playable replicas of real trading-firm OAs (Optiver, SIG, IMC).
C++20 implementation of special-relativistic geometry applied to OHLCV data: Lorentz factors, spacetime intervals, Christoffel symbols, and geodesic deviation signals from live market data. DOI: 10.5281/zenodo.18639919.
An library to price financial options written in Python. Includes: Black Scholes, Black 76, Implied Volatility, American, European, Asian, Spread Options.
Implementation of code snippets, exercises and application to live data from Machine Learning for Asset Managers (Elements in Quantitative Finance) written by Prof. Marcos López de Prado.
Functions, examples and data from the first and the second edition of "Numerical Methods and Optimization in Finance" by M. Gilli, D. Maringer and E. Schumann (2019, ISBN:978-0128150658).
Implement, demonstrate, reproduce and extend the results of the Risk articles 'Differential Machine Learning' (2020) and 'PCA with a Difference' (2021) by Huge and Savine, and cover implementation details left out from the papers.
This repository contains codes that were executed during my training in the CQF (Certificate in Quantitative Finance). The codes are organized by class, facilitating navigation and reference.
An opinionated approach to empirical research in financial economics - a fully transparent, open-source code base in multiple programming languages (Python and R) to enable the reproducible implementation of financial research projects for students and practitioners.
Open-source backtesting simulator with a 25-minute Assumption Clinic worksheet that teaches students to inspect backtest assumptions using published failed cases.
Free Korean equity settled closes with 250 trading days of per-stock history, DART filings and earnings as JSON/CSV, without signup, an API key or request quotas.
Point-in-time US equity fundamentals from SEC EDGAR with filing dates and restatement flags; free API access includes 250 requests per day without a credit card, and a 40-company CC0 annual dataset is available without signup.
Financial market data API with a permanent free personal-use tier offering two years of US stock history, end-of-day prices, and reference data at five requests per minute, with no credit card required.
Search, social, commerce, and news-sentiment data for alternative-data research, with a permanent free tier of 100 requests/month, 90 days of history, and top-10 trend boards delayed 24 hours.
Financial data API for quantitative research, with a permanent free tier of 1,000 monthly API calls and one year of historical data, without a credit card. [PyPI](https://pypi.org/project/axionquant-sdk/).
Free web app for crypto liquidation price, position sizing, PnL and funding-rate calculations across major exchanges, with documented formulas on the site's Learn hub.
Proprietary local backtesting and optimization CLI with a permanent free plan requiring no registration, limited to historical data through 2023-12-31; Pine Script export requires a paid plan.
Forex trading and market-data gateway for MetaTrader 4/5 and OANDA, with a free plan for one account, 500 OHLC requests and 100 trades per month, without a credit card.
Live autonomous benchmark that evaluates LLM trading performance on DAX and Nasdaq indices using identical strategies and real-time market data. API access available.
Agent-verification platform with EdgeProof backtest diagnostics using Deflated Sharpe, permutation tests, and out-of-sample decay, offering five free checks per day without signup or payment information.
Financial-text classification API for sentiment, directionality, quality, post type, relevance, confidence, and sarcasm, with 1,000 free credits per month and no credit card required.
SEC financial-statement analysis and API with a permanent free account covering five selected stocks and five years of data, without payment information.
Brazilian stock and FII data API with fundamentals, prices, dividends, financial statements, and macroeconomic data, offering 200 free requests per day without a credit card.
Brazilian market data API with a free tier of 15,000 requests per month and up to three months of price history, plus four stocks accessible without an account.
Brazilian stock and FII analysis with free essential fundamentals, one year of price history, Bazin and Graham valuations, and the latest CVM statements, without a credit card.
Public SEC ownership research web app for exploring 13F portfolios, insider activity, Schedule 13D/G filings, company facts, and latest filing workflows.
SEC filings, insider transactions, and institutional holdings API with a permanent free tier of 5,000 requests per month and 15 requests per minute, without a credit card.
SEC filing events as typed JSON: 8-K item codes with materiality flags, SC 13D/13G activist stakes (holder, target, percent of class), merger forms, and S-1/424B IPO filings, polled over REST or pushed via HMAC-signed webhooks, sourced from data.sec.gov.
Real-time SEC EDGAR filings normalized to one JSON schema: Form 4 insider trades, 8-K corporate events, and S-1/IPO registrations, delivered via REST, HMAC-signed webhooks, and a hosted MCP server, with a permanent free tier (2,500 req/mo, full schema).
Financial data API with a free personal-use plan allowing 300 requests per day for symbol lists and selected market data, including historical commodity and OTC prices.
Free SEC 13F research web app covering 11,700+ institutional managers and 208,000+ filings from 2019 onward. Quarter-over-quarter position changes with share-based thresholds, confidential-treatment reveals badged, per-filing links to the EDGAR source document, and Excel export on every table. No account required.
Financial press-release platform covering Business Wire, PR Newswire, GlobeNewswire, and AccessWire, with a permanent free Wire dashboard delayed five minutes and no credit card required.
Portfolio analysis and optimization API with anonymous free access to selected endpoints for up to 20 assets, subject to a shared one-request-per-second limit.
Prediction-market data service with free market metadata and tag datasets without an account, while tick-level data requires paid access after five trial file downloads.
Currency exchange rates, conversion and VAT data with a permanent free tier of 200 requests per day and no credit card; historical data require a paid plan.
Market data over REST and WebSocket with a permanent free non-commercial tier per market: 10,000 monthly REST calls, 60 requests per minute, five streaming symbols and one month of history, without a credit card.
Hyperliquid order-flow-enriched minute OHLCV with a permanent free tier covering BTC, ETH, SOL and HYPE over the trailing month, with unlimited Parquet downloads and no credit card.
Technical, quantitative and sentiment data for stocks and ETFs with risk metrics, peer percentiles and market regime signals. Optimized for AI/LLM agents.
Trade-review workspace with a permanent free tier for 25 trades per month, CSV imports and P&L, win-rate and R-multiple analytics without payment information.
Institutional-grade financial intelligence platform. Track 43K+ congressional trades (STOCK Act), SEC insider Form 4 filings, 25 superinvestor 13F portfolios, CFTC futures positioning, ARK ETF holdings, and short interest — all scored by AI for signal strength.
Investment decision frameworks with a permanently free library of investor principles and scenarios covering valuation, position sizing and risk discipline.
Cryptocurrency research terminal with free access without signup and a personal API tier allowing 10,000 requests per month and 30 requests per minute.
Aggregates public financial data from HKEX, the SFC, the Hong Law Society, UK Companies House and other sources, has searchable datasets on listed companies, many in machine-readable formats.
Alternative-data research on private companies using GitHub commit velocity and contributor growth, with a permanently free weekly digest covering five startups and basic MCP read tools.
REST API for US equities & options: reference & fundamental data, multi-year financial statements, corporate events, analyst consensus, a screener, and order execution.
Trading strategy builder that turns plain-English market ideas into inspectable rules, backtests them against historical data, and monitors matching live conditions.
AI forecasting-accountability lab: every forecast is locked pre-event, Bitcoin-timestamped (OpenTimestamps), and Brier-scored against prediction markets in public.
Trading-signal validation service with a free browser-based Deflated Sharpe Ratio calculator requiring no account, alongside paid backtest validation and forensics.
Daily trend, momentum, correlation and risk analytics with a permanently free stock and ETF terminal, while API exports and cointegration screening require paid access.
Market statistics and options-positioning analytics with free public max-pain and GEX pages and 10 API calls per UTC day using an email-only key, without payment information.
Hourly crypto news sentiment and typed market events (hack, regulation, listing, delisting, legal) for 30 pairs, served as point-in-time buckets whose SHA-256 hash is committed to a public append-only log at publish time, so a backtester can check the history was never rewritten. Paid plans are $24, $79 and $249 a month before VAT; the permanent free tier is 1,000 requests a day with BTC, ETH and SOL live, the other 27 pairs delayed 24h, 1 day of history depth, no card, non-commercial use.
Hyperliquid and Lighter market data through REST, WebSocket and replay, with a permanent free tier offering 50,000 credits per month, 15 requests per second and the most recent 30 days of history without a credit card.
Forex news sentiment with a permanent free API allowing 25 requests per day, one three-hour cycle of delay and non-commercial use with attribution; no credit card or historical API series. [PyPI](https://pypi.org/project/fxnewsbias/).
Hosted MT4 and MT5 broker API with historical candles, real-time tick streaming, and a permanent free tier for five demo broker accounts and unlimited demo orders, without a credit card; live-account automation is paid. [Docs](https://tickerall.com/docs).
Derived, indicative multi-asset market data through REST, WebSocket, and MCP, with a permanent free tier of 3,000 REST requests per month and no credit card required. [Docs](https://tickerlayer.com/docs).
Monthly statistics relating proprietary SEC-filing importance scores to next-session excess stock moves, with per-event data, reproduction metadata, and a CC BY 4.0 license.
Natural-language trading-strategy builder with backtesting and forward testing; a permanent free tier includes five backtests and 40 assistant messages per month without payment details, while live trading requires a paid plan.
Stock research terminal with free access to facts, standard ratios, universe-wide screening, 13F top holdings, three saved portfolios, a 25-name watchlist and three-year backtests; Pro adds valuation and full portfolio analytics.
Free live open/closed clock and reference for 69 futures products across 14 exchanges with DST-aware session windows, night sessions, 2026 holiday calendars, contract specifications sourced from official exchange publications, bilingual EN/ZH, and an open CORS JSON endpoint.
Cross-chain swap and bridge route comparison that queries exchange providers for a given pair and amount and ranks the routes by output, with rate, fees, ETA and price impact per route; free to use with no account, no API key and no added fee. [Docs](https://eidex.io/docs).
Cryptocurrency market alerts, screeners, and funding-rate, liquidation, and chart-pattern boards, with a permanent free tier of five active alerts and ten notifications per day without a credit card; public research boards require no account.
Financial and geopolitical news API with a permanent free REST tier of 5,000 requests per month, a 12-hour delay and one month of history, without a credit card.
Brazilian company and real-estate fund fundamentals with public calculation methodology and a permanent free API tier of 200 requests per day without a credit card.
Bitcoin and macroeconomic market briefings through REST and MCP, with a permanent free tier of six briefings, 30-day history, 60 requests per hour and 500 per day, without a credit card.
Factor data and cosine-similarity screening with a permanent free API tier of 250 calls per day, daily factor rankings and market-context snapshots, without a credit card.
Bitcoin cycle scores and market-state indicators through four free public API endpoints requiring no account or key, with responses cached for 5–60 minutes.
Public earnings-call transcripts and an earnings calendar readable without an account, with paid API access for programmatic transcripts, audio and slide decks.
Macroeconomic release data with permanent free access to USD announcements covering the latest 90 days at 100 requests per day without an API key or payment information.
Free market and economic data with up to 10 databank downloads per hour of one million rows each, subject to a shared monthly bandwidth allowance, without a credit card.
Archived [protocol implementation](https://github.com/rainx/pytdx/tree/master) retained for studying TongDaXin binary market-data protocols and local quotation-file formats.