Walk Forward Validation
Walk-forward validation framework for trading strategies and ML models with time-series-aware splits, overfit detection, and regime-aware validation
Install
/plugin marketplace add agiprolabs/claude-trading-skillsThe repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.
npx skills add agiprolabs/claude-trading-skills --skill walk-forward-validationgit clone --depth 1 https://github.com/agiprolabs/claude-trading-skills.git /tmp/claude-trading-skills
mkdir -p ~/.claude/skills
cp -r /tmp/claude-trading-skills/skills/walk-forward-validation ~/.claude/skills/walk-forward-validationPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Walk-Forward Validation Walk-forward validation framework for trading strategies and ML models. Standard cross-validation (k-fold, random splits) fails catastrophically for financial time series because it introduces lookahead bias and ignores autocorrelation. This skill covers proper time-series validation techniques including rolling and expanding windows, purged cross-validation, combinatorial purged cross-validation (CPCV), and overfit detection metrics. ## Why Standard Cross-Validation Fails Standard k-fold CV assumes data points are independent and identically distributed (IID). Financial time series violate both assumptions: 1. **Lookahead bias** — Random splits let the model train on future data and predict past data, artificially inflating performance. 2. **Autocorrelation** — Adjacent observations are correlated. A random split that puts Monday in test and Tuesday in train leaks information. 3. **Regime dependence** — Markets shift between regimes. A model trained on a bull market and tested on a bull market tells you nothing about bear market performance. 4. **Label overlap** — If labels are computed over windows (e.g., 24h forward return), adjacent train/test samples share label computation periods, leaking information. ## Walk-Forward Framework ### Rolling Window (Fixed Train Size) The train window has a fixed size and slides forward in time. This is preferred when you believe older data is less relevant (common in crypto).
Details
- Skill name
walk-forward-validation- Repository
- agiprolabs/claude-trading-skills
- Path
skills/walk-forward-validation/SKILL.md- License
- MIT
- Repo updated
- 3 September 2026
FAQ
What does the Walk Forward Validation skill do?
Walk-forward validation framework for trading strategies and ML models with time-series-aware splits, overfit detection, and regime-aware validation
How do I install the Walk Forward Validation skill?
Claude Code plugin: /plugin marketplace add agiprolabs/claude-trading-skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Walk Forward Validation popular?
It lives in agiprolabs/claude-trading-skills, which has 564 GitHub stars and 113 forks.
Related skills
Backtrader
564agiprolabs/claude-trading-skills
Event-driven backtesting with bar-by-bar execution, complex order types, multiple analyzers, and custom indicators
Birdeye API
564agiprolabs/claude-trading-skills
Solana token market data via Birdeye — prices, OHLCV, trades, token metadata, security checks, and trader activity
Coingecko API
564agiprolabs/claude-trading-skills
Broad crypto market data from CoinGecko covering 13,000+ tokens. Global market stats, historical price data going back years, exchange volumes, trending tokens, and category filters. Best for macro analysis and long-term historical data.
Agent Memory MCP
514Dokhacgiakhoa/Agent-Skills-4-Vibe-Coding-CLI
A hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).
Volatility Modeling
564agiprolabs/claude-trading-skills
Volatility estimation, forecasting, and regime classification using GARCH, EWMA, realized volatility, and volatility cones
03 Context Generate
564ai-driven-dev/framework
Route a request to generate a context artifact (skill, rule, agent, command, or hook) to its generator when the kind is unnamed. A named kind triggers its generator directly. Not for listing existing artifacts.
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.