A股个人择时选股系统 — 三级择时漏斗 × 财务排雷 × 量化策略引擎 | FastAPI + Next.js 16 + SQLite

The significance test for Granger causality of financial risks in stock indices

A forward-looking risk prediction system that analyzes 31 multi-dimensional signals across news, finance, operations, climate, and geopolitics to predict supplier disruption probability 3–6 months in advance.

🏦 Machine Learning system for credit default prediction using a RandomForestClassifier. Features an end-to-end pipeline including synthetic financial data generation, robust preprocessing (ColumnTransformer), and comprehensive evaluation with ROC-AUC and Confusion Matrices.

Batch financial risk scoring and Isolation Forest anomaly detection project using Python, SQL, and BI-ready outputs.

Macro-financial stress testing framework projecting UK bank capital under baseline and adverse macroeconomic scenarios.

Autonomous financial risk intelligence platform, SEC EDGAR ingestion, Isolation Forest anomaly detection, Benford's Law forensic analysis, corporate network graph, SHAP explainability, and Gemini AI analyst. Built with DuckDB, dbt, and Streamlit.

End-to-end MLOps pipeline predicting corporate bankruptcy from 95 financial ratios. Features nested CV, SHAP explainability, MLflow tracking, and Dockerized FastAPI deployment.

Independent validation framework for CCP-style initial margin models, including VaR, margin add-ons, backtesting, stress testing, sensitivity analysis, procyclicality monitoring, and model-risk governance.

CashAegis is an autonomous cashflow control-plane research prototype for forecasting, risk signals, policy-governed action drafts, and audit evidence. Execution is simulated only; no real financial or compliance actions.

Intelligent EMI prediction & financial risk assessment platform with dual ML models (Classification + Regression) and MLflow experiment tracking

ML pipeline combining ASX market data & macroeconomic indicators for financial risk detection. FastAPI + Streamlit + XGBoost.

Data engineering pipeline for financial risk AI — feature engineering, SEC EDGAR ingestion, synthetic fraud generation, and LLM instruction pair export. Part of the IntelligenceOS build.

bankALM — Public-data U.S. bank fragility atlas. Transparent deposit-run, ALM-mismatch, and composite fragility scores for 5,000+ FDIC-insured banks.

SQL-based credit risk analytics project using public Home Credit data, DuckDB, data quality checks, customer-level feature engineering and rule-based risk segmentation.

AI-powered corporate bankruptcy risk prediction system using financial ratios, XGBoost, SHAP explainability, and multi-year financial distress forecasting.

Course project scaffold for China Railway financial risk analysis with risk graph, text indicators, and warning model

End-to-end risk data engineering pipeline with SQLMesh, DuckDB, Python analytics, audits, and financial risk metrics.

Financial risk SDKs for AI agents — Python and TypeScript clients plus LangChain, CrewAI, LlamaIndex and Vercel AI SDK adapters for portfolio risk, crypto risk and market verdicts.

Machine learning-based bankruptcy prediction and credit risk decision support system developed with XGBoost, SHAP, and Streamlit.