Walk forward analysis
Skill 0xgetz/xi-agent-skills/trading-skills/walk-forward-analysis
Analyze markets using Walk Forward Analysis (out-of-sample test). Validate strategies with rolling in/out-of-sample windows. Activate when the user asks to analyze, interpret, or build signals based on Walk Forward Analysis or out-of-sample test.From its SKILL.md
npx -y skills add 0xgetz/xi-agent-skills --skill walk-forward-analysisAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.8 KB, 346 tokens by cl100k_base, as published. Nobody here has run it
Walk Forward Analysis
Overview
Walk Forward Analysis (out-of-sample test) — Validate strategies with rolling in/out-of-sample windows.
When to use this skill
Use when the user asks to:
- Analyze or interpret Walk Forward Analysis on a chart or dataset
- Build buy/sell signals or alerts based on out-of-sample test
- Combine Walk Forward Analysis with other indicators for confirmation
How it works
Validate strategies with rolling in/out-of-sample windows. Apply it on OHLCV data (open, high, low, close, volume) for any timeframe. Always confirm with price structure, trend context, and at least one independent indicator before acting.
Reading the signals
- Bullish bias: signal aligns with higher highs/higher lows and rising volume.
- Bearish bias: signal aligns with lower highs/lower lows and rising volume.
- No-trade: conflicting context or low volatility/volume.
Worked example (Python)
import pandas as pd
# df has columns: open, high, low, close, volume (datetime index)
# Compute the indicator, then generate signals
# (use pandas/numpy or ta libraries; validate on out-of-sample data)
Risk management
- Define stop-loss from structure or ATR before entry.
- Size positions by fixed-fractional risk (e.g. 0.5–1% per trade).
- Never rely on a single indicator; require confluence.
Common pitfalls
- Over-optimizing parameters to past data (curve fitting).
- Ignoring the higher-timeframe trend.
- Acting on signals during low liquidity.
Educational analysis only. Not financial advice.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.