🦞 AI 量化交易系统 — 37 因子选股 · 8 层风控 · 同花顺数据大屏 · 全自动盯盘

A research project on macro regime clustering using a compact monthly feature space and Jump Models, with stability analysis, external validation, and asset mapping across equities, bonds, oil, and gold.

active investing

An end-to-end machine learning trading system: ensemble of transformer models, hybrid RNN model and LightGBM with temperature calibration, and a live trading bot with Kelly Criterion position sizing.

Machine learning overlay for SPY using ^GSPC regime signals, Jump Model labels, and XGBoost. Focused on downside protection, recovery timing, and risk-adjusted improvement under a unified long-sample backtest.

Identify regimes in financial markets based on multivariate time series data using multiple methodologies, including CNN, AutoEncoder, Siamese Model, Correlation Matrices, K-means++, and Hierarchical Clustering

Free REST API for the moneyfeel Macro & Geopolitical Risk Index (MRI) — market regime classifier covering 5 regions across Daily/Weekly/Monthly timeframes. Free for researchers, quants and portfolio managers.

An institutional-grade, asynchronous High-Frequency Trading (HFT) node built in Python. This system reconstructs Level-2 (L2) Limit Order Book (LOB) updates, computes latency-critical microstructure features in O(1) time, and uses a Gaussian Hidden Markov Model (HMM) to infer latent market regimes.

Detects financial market regimes using semantic geometry of news embeddings (FinBERT) with clustering and statistical modeling to predict volatility and regime shifts ahead of price action.

Reproduced a research-backed Wasserstein k-means framework for market-regime detection by clustering rolling return windows as empirical probability distributions, implementing 1D optimal-transport distances, Wasserstein barycenters, baseline comparisons, validation metrics, and regime-aware visualization/backtesting in Python.

MacroPulse is a real-time macro intelligence platform that transforms fragmented global data into actionable investment insights through a unified decision layer.

Institutional-grade market regime decoding via Savitzky-Golay Kinematics and Hidden Markov Models (HMM). Engineered for zero-lag signal demodulation and structural alpha detection.

📊 Detect market regimes by clustering probability distributions using Wasserstein K-means for more accurate financial analysis and insights.

Multi-model framework for market regime detection and dynamic asset allocation using HMM, XGBoost, LSTM, backtesting, and RL.

Lightweight pip-installable market-regime classifier (LATERAL / STRONG_TREND / VOLATILE) with GMM, HMM, and rule-based methods.

Dual-channel regime detector: Spectral Gap (Landon-Xian 2025) + Geometric Observables (Hammond 2026) from eigen-spectra. Entropy, purity, eigenvector dynamics combined (Cohen’s d~0.7). Practice.

Is the crypto market bull, chop, or bear? Market regime label + per-coin RUN gate from public daily klines. No API keys, zero dependencies.

Deterministic market regime detection agent with persistent state and transparent signal-based classification.

Multi-layer market regime classifier — BTC trend (BULL/BEAR/CHOPPY/TRANSITION), GOLD macro (RISK_ON/OFF), contrarian Fear & Greed

BTC/USDT algo trading: 4-state regime classifier + Kalman filter + Hurst exponent achieves Sharpe >6, max drawdown <15%, beat benchmark in 12/17 quarters