<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> This project provides a Markov regime detection skill for financial assets, enabling users to understand market states (Bull/Bear/Sideways), their persistence, and associated risks. It targets quantitative traders and developers seeking to integrate sophisticated market regime analysis into existing strategies or workflows without complex setup, offering a composable, easy-to-install tool that enhances trading decisions with data-driven insights.

How It Works

The framework classifies daily regimes (Bull/Bear/Sideways) via a rolling-return rule (default 20-day, ±5%). It constructs a 3x3 transition matrix from historical data using maximum-likelihood estimation, forecasting future regimes via Chapman-Kolmogorov. A stationary distribution reveals the long-run regime mix, and a signed signal (bull_prob - bear_prob) indicates direction/conviction. An optional Hidden Markov Model (HMM) fit using hmmlearn provides alternative analysis, with graceful degradation if compilation fails. The design prioritizes composability, allowing seamless integration into existing trading agents or data pipelines as a confirmation layer or risk filter.

Quick Start & Requirements

Installation is streamlined via a two-command plugin process for Claude Code: /plugin marketplace add jackson-video-resources/markov-hedge-fund-method /plugin install markov-hedge-fund-method@markov-hedge-fund-method A "zero-trust" manual build involves pasting a script into Claude Code's