advanced-math-tradingoptimization-advanced

Advanced optimization for trading: OCO, multi-objective, DRO, MIP, Bayesian opt, plus convex/Markowitz/BL references.

⭐1stars🔀0forks📁View on GitHub🕐Updated Dec 28, 2025⬇️1downloads

When & Why to Use This Skill

This Claude skill provides a comprehensive suite of advanced mathematical optimization techniques specifically designed for quantitative trading and financial modeling. It enables the implementation of sophisticated strategies using Online Convex Optimization (OCO), Multi-objective optimization (NSGA-II), and Distributionally Robust Optimization (DRO). By bridging classic frameworks like Markowitz and Black-Litterman with modern Mixed-Integer Programming (MIP) and Bayesian optimization, it empowers users to build robust, risk-aware, and highly efficient trading portfolios.

Use Cases

  • Dynamic Portfolio Rebalancing: Use Online Gradient Descent (OGD) and Follow the Regularized Leader (FTRL) snippets to create adaptive strategies that respond to real-time market shifts.
  • Multi-Objective Trade-off Analysis: Balance competing financial metrics such as return, risk, and turnover using the NSGA-II algorithm for optimal decision-making.
  • Tail-Risk Protection: Implement Distributionally Robust Optimization (DRO) with CVaR or moment-based constraints to ensure portfolio robustness against extreme market volatility.
  • Real-World Constraint Modeling: Apply Mixed-Integer Programming (MIP) to handle practical trading limitations like cardinality constraints, round lots, and transaction costs.
  • Strategy Hyperparameter Tuning: Leverage Bayesian optimization to efficiently find the best-performing parameters for complex algorithmic trading signals.
nameadvanced-math-trading/optimization-advanced
descriptionAdvanced optimization for trading: OCO, multi-objective, DRO, MIP, Bayesian opt, plus convex/Markowitz/BL references.

What this covers

  • Online convex optimization, multi-objective (NSGA-II), DRO (CVaR/moment), mixed-integer constraints, Bayesian optimization.
  • Links to convex/Markowitz/Black-Litterman background.

Navigation (load on demand)

  • docs/knowledge-base/domains/foundations/advanced-mathematics/advanced-optimization.md — main code (OGD/FTRL, NSGA-II, DRO cvxpy, MIP gurobi, Bayesian opt).
  • docs/knowledge-base/domains/foundations/advanced-mathematics/convex-optimization.md
  • docs/knowledge-base/domains/foundations/advanced-mathematics/optimization.md
  • docs/knowledge-base/domains/foundations/advanced-mathematics/mean-variance-optimization-markowitz.md
  • docs/knowledge-base/domains/foundations/advanced-mathematics/black-litterman-model.md

Quick workflows

  • Adaptive strategy/portfolio → use OGD/FTRL snippets.
  • Return/risk/turnover trade-offs → NSGA-II block in advanced-optimization.
  • Tail/robustness → DRO CVaR or moment-based cvxpy blocks.
  • Cardinality/round lots → MIP section (gurobi).
  • Hyperparameter tuning → Bayesian opt section.

Notes

  • Choose the minimal subset of files; avoid loading everything.