skills.rest
dnkrow
dnkrowCOMMUNITY
Explore all skills by this author →
GitHub

volatility-modeling

Estimate crypto volatility and classify regimes using econometric estimators and percentile cones.

1|Updated May 15, 2026
Search Tags:#risk management#garch#ewma#volatility estimation#volatility cones#crypto markets
One-click install
npx skills add https://github.com/dnkrow/skill --skill volatility-modeling-dnkrow
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: volatility-modeling
Source: https://github.com/dnkrow/skill/tree/main/claude-global/volatility-modeling
Command: npx skills add https://github.com/dnkrow/skill --skill volatility-modeling-dnkrow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires numpy, pandas, scipy, httpx, and includes scripts (resource) and references (resource) components.

What problem does it solve?

It helps you quantify how volatile a crypto asset has been and what volatility regime it is in, so you can make better risk and strategy decisions instead of relying on intuition.

Core Features & Use Cases

  • Estimate volatility from price data: Computes realized volatility using close-to-close, Parkinson (high-low), Garman-Klass (OHLC), EWMA, and GARCH(1,1).
  • Forecast future volatility: Produces forward-looking volatility using EWMA’s responsiveness and GARCH’s term structure (multi-horizon variance forecasts).
  • Detect volatility regimes with context: Builds volatility cones (percentiles across multiple lookback windows) and classifies current volatility into low/normal/high/crisis regimes.
  • Use Case: For a trading workflow, estimate current annualized volatility for a token, determine whether conditions are compressed or elevated via cone percentiles, and choose sizing/stop/risk posture consistent with the inferred regime.

Quick Start

Use the volatility-modeling skill to compute volatility estimates, a volatility cone, and a regime label for the token from 400 days of OHLC data.

Frequently Asked Questions about volatility-modeling

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

FAQPage Schema
How do I estimate and forecast crypto volatility using historical OHLC data?▼

You can estimate crypto volatility by computing realized volatility from log returns or OHLC ranges using close-to-close, Parkinson, and Garman-Klass estimators, then forecasting multi-horizon volatility with EWMA and GARCH models.

What is a volatility cone and how does it classify volatility regimes?▼

A volatility cone builds rolling percentile distributions of volatility across multiple lookback windows to classify current conditions into low, normal, high, or crisis regimes, providing statistical context for risk budgeting.

Can I use GARCH and EWMA to calculate forward-looking volatility for position sizing?▼

Yes, you can generate forward-looking volatility forecasts using EWMA's responsiveness and GARCH's term structure to calibrate regime-aware position sizing, stop placement, and strategy selection for crypto assets.

Does this volatility modeling approach work with Birdeye OHLCV data for crypto markets?▼

Yes, it processes OHLC history from demo or live Birdeye OHLCV data, applying close-to-close, Parkinson, Garman-Klass, EWMA, and GARCH(1,1) estimators to calculate annualized volatility for crypto tokens.

What is the best way to detect if a crypto token is in a compressed or elevated volatility regime?▼

Build a volatility cone using percentile distributions over multiple lookback windows to classify the token's current volatility regime, enabling you to adjust your risk posture and trading strategy accordingly.

Why use multiple volatility estimators instead of just close-to-close returns?▼

Using multiple estimators like Parkinson and Garman-Klass captures intraday high-low ranges, while EWMA and GARCH add term-structure diagnostics, yielding more robust annualized volatility estimates than close-to-close alone.

Related Skills

Explore More Finance & Accounting Skills→
ebrahim-saniebrahim-sani

smc

Detect BOS, ChoCH, FVG, and order block signals from OHLCV pandas DataFrames.

Community
Intermediate
Drakkar-SoftwareDrakkar-Software
6.3k

trading-strategy

Synthesize technical analysis, sentiment, and risk management into trading recommendations.

Official
Advanced
JoelLewisJoelLewis
164

diversification

Calculate portfolio variance, volatility, diversification ratio, HHI concentration, and risk contributions from covariance matrices and asset weights.

Community
Intermediate
silversteinsilverstein
21

aeon

Build time-series machine learning models with scikit-learn compatible APIs.

Community
Intermediate