Agentic coding skills for backtesting trading strategies using VectorBT. Supports Indian, US, and Crypto markets with realistic transaction cost modeling, TA-Lib indicators, QuantStats tearsheets, and 12 ready-made strategy templates.

An implementation of Giuseppe Paleologo's Rademacher Antiserum, designed to assess strategy performance consistency through Rademacher complexity and RAS-adjusted Sharpe Ratios. This code evaluates strategy robustness by applying Rademacher random vectors for anti-overfitting analysis.

Crypto-currencies (Bitcoin) trading & Transfer Learning applications

Helix Treasury System

Utilized time series, statistic, ML and NLP models to practice. Topics include stock forecasting (algorithm trading), US 2024 presidential election and customer sentimental review.

Method to use rolling optimization with 3D LSTM arrays. Explores integrating multi-asset portfolio with embedding layer.

Quant research toolkit: 4-asset equal-weight risk-parity basket (BTC/SP/TLT/GLD) ships as production after 5 rounds of adversarial review. Includes non-overlap sampler, multi-seed evaluator, DSR, bootstrap, pristine holdout, cost sensitivity. Markov-chain BTC strategy retained as sunset internals. Experimental research, not financial advice.

SiraEdge Portfolio Optimization Module - Complete implementation with 7 optimization models, ML integration, and comprehensive documentation

AI-driven hybrid LightGBM–Ridge model for dynamic financial time-series allocation with Streamlit UI.

Temporal cross-validation for insurance pricing models. Walk-forward splits respecting policy/accident year structure and IBNR development buffers.

Intraday mean‑reversion trading strategy with realistic execution costs, turnover caps, and walk‑forward backtesting.

End-to-end quantitative trading system for Taiwan stocks — DoubleEnsemble model, 300+ factors, Walk-Forward backtesting, React dashboard

Top Indian Stock Market Analysis Tools 2026 📈 | Free NSE/BSE, Derivatives & Trading AI Hub

Oil Production Flow Rate Prediction with Deep Neural Network Algorithm such as Recurrent Neural Network (RNN) and Long-Short Term Memory (LSTM). This Model will testing with Validation method called Walk-Forward Validation (WFV). Basically the validation seperate in two part, the first WFV over actual data and WFV over predicted data.

Temporal cross-validation for insurance pricing - respects policy time structure, CatBoost, Polars

Enable Claude to analyze Indian equity markets with modules covering NSE/BSE stocks, derivatives, flows, news, and trade planning.

Reproducible regime-aware multi-strategy trading system for BTC/USDT, with walk-forward validation and stress-tested risk controls.

Provide comprehensive backtesting skills for trading strategies across Indian, US, and crypto markets using VectorBT and AI coding agents.