symbol-selection-statistical

Select trading symbols using Hurst, half-life, GARCH, and regime-persistence metrics.

3|Updated Dec 10, 2025
One-click install
npx skills add https://github.com/smith6jt-cop/Skills_Registry --skill symbol-selection-statistical
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: symbol-selection-statistical
Source: https://github.com/smith6jt-cop/Skills_Registry/tree/main/plugins/trading/symbol-selection-statistical/skills/symbol-selection-statistical
Command: npx skills add https://github.com/smith6jt-cop/Skills_Registry --skill symbol-selection-statistical

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Selecting assets for predator-prey Markov trading systems requires advanced statistical metrics beyond simple screens.

Core Features & Use Cases

  • Hurst exponent analysis (short-term <0.45) for mean-reversion and long-term >0.55 for trend.
  • Half-life of mean reversion to gauge trading horizon.
  • GARCH fit quality scoring to ensure volatility model adequacy.
  • Regime persistence via HMM to assess regime duration and stability.
  • Use case: combine metrics to select assets with clear regime separation and robust volatility modeling.

Quick Start

Define calculate_hurst_dfa, calculate_half_life, score_garch_fit, and calculate_regime_persistence. Run on a price series to obtain selection scores and regime metrics.

Frequently Asked Questions about symbol-selection-statistical

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I select trading symbols based on mean-reversion and momentum compatibility?

Symbol selection for mean-reversion and momentum strategies requires statistical screening using Hurst exponent analysis, half-life estimation, and GARCH volatility modeling. This Skill computes short-term Hurst metrics (<0.45 for mean-reversion, >0.55 for trend), mean-reversion half-life, and GARCH fit quality to identify assets with clear regime separation and robust volatility dynamics across equities, futures, FX, and cryptocurrencies.

What is a Hurst exponent and how does it help with asset selection?

The Hurst exponent measures market behavior on a scale of 0 to 1, with values below 0.45 indicating mean-reverting price patterns ideal for range-trading strategies, and values above 0.55 indicating trending behavior. This Skill calculates both short-term and long-term Hurst metrics to classify assets and filter those compatible with your trading regime.

Can I use regime detection with Markov models to screen assets?

Yes. This Skill integrates HMM-based regime persistence analysis to detect predator-prey Markov regimes, measuring regime duration and stability. It scores assets on regime persistence and other statistical dimensions, enabling you to select symbols with well-defined regime separation suitable for regime-aware trading strategies.

How do I assess volatility model fit when selecting trading symbols?

GARCH fit quality scoring evaluates whether a GARCH volatility model adequately captures an asset's risk dynamics. This Skill computes GARCH diagnostics as part of its weighted composite scoring, ensuring selected symbols have volatility patterns that GARCH models can reliably forecast for downstream risk management.

What data inputs and time horizons does this selection process support?

The Skill applies to asset universes across equities, futures, FX, and cryptocurrencies at intraday to daily horizons. It accepts price series as input and outputs selection scores, Hurst metrics, half-life estimates, regime persistence statistics, and GARCH diagnostics via a configurable weight schema.

What's the best way to combine multiple statistical metrics for asset screening?

This Skill provides a weighted composite scoring workflow that integrates Hurst exponent, half-life, GARCH fit quality, and regime persistence into a single selection score. You define the weight schema to prioritize metrics matching your strategy, then run the pipeline on your price series to rank and filter compatible symbols.