Validate quantitative trading signals with walk-forward backtesting and deflated Sharpe ratio.
npx skills add https://github.com/tmcga/alpha-stack --skill quant-signals
Please help me install this Agent Skill. Skill: quant-signals Source: https://github.com/tmcga/alpha-stack/tree/main/skills/quant-signals Command: npx skills add https://github.com/tmcga/alpha-stack --skill quant-signals
Quantitative signal design and testing is often ad-hoc, error-prone, and lacks rigorous out-of-sample validation. This Skill provides a structured framework to define, test, and validate trading signals with robust statistical checks, backtesting discipline, and ensemble construction to improve durability.
Design a new alpha signal for a chosen universe, specify data sources, apply a normalization and neutralization pipeline, run a walk-forward backtest, and interpret the out-of-sample results considering costs.
High-intent search queries and answers about installing and using this skill.
Walk-forward optimization validates quantitative trading signals by testing parameters on unseen out-of-sample data. This Skill implements it alongside CPCV cross-validation and deflated Sharpe ratio adjustments to prevent overfitting and evaluate true performance.
You design an alpha signal by formalizing hypotheses and transforming raw data into testable signals. This Skill lets you apply normalization, neutralization, and walk-forward backtesting across multi-asset universes with realistic transaction cost modeling to evaluate out-of-sample performance.
Yes, cross-sectional IC evaluates the predictive correlation of trading signals across assets. This Skill emphasizes cross-sectional Information Coefficient analysis to validate signal strength before combining multiple signals in an ensemble.
Yes, ensemble construction combines multiple alpha signals to improve durability. This Skill lets you manage signal correlation and calibrate conviction based on agreement levels, optimizing portfolio allocation under risk constraints.
The deflated Sharpe ratio adjusts performance metrics for multiple testing bias to prevent false discoveries. You need it during backtesting to ensure quantitative signals are not statistical flukes, and this Skill integrates it as a core validation requirement.
Machine learning risks overfitting and model drift in quantitative signal design. This Skill mitigates these limitations by integrating ML components with strict guardrails, walk-forward optimization, and CPCV cross-validation to maintain out-of-sample predictive validity.
cx-qkTransform SHFE, DCE, CZCE, and GFEX futures data into unified Excel templates.
PatrickSUDOFetch live brokerage positions and generate portfolio reports with sector allocation, P&L, options, and risk.
snoodleboot-ioMonitor machine learning model performance and detect prediction or data drift.
JoelLewisCompute optimal portfolio weights using mean-variance optimization and Black-Litterman models.
