Run ADF, Hurst, and variance ratio tests to assess mean reversion.
npx skills add https://github.com/dnkrow/skill --skill mean-reversion-dnkrow
Please help me install this Agent Skill. Skill: mean-reversion Source: https://github.com/dnkrow/skill/tree/main/claude-global/mean-reversion Command: npx skills add https://github.com/dnkrow/skill --skill mean-reversion-dnkrow
Mean-reversion analytics determine whether a price (or spread) series is statistically likely to revert to a long-run average, so you can build trading signals with measurable evidence rather than guesswork.
Use Case: You suspect a crypto token is mean-reverting after a shock; you run the analysis to confirm stationarity and estimate half-life, then use the z-score framework to decide whether conditions are suitable for a mean-reversion long/short and what horizon to use.
Run the mean-reversion demo analysis using synthetic OU data to see a full report of ADF/Hurst/variance ratio/half-life/OU parameters and the current z-score signal for the series.
High-intent search queries and answers about installing and using this skill.
Mean reversion testing uses ADF, Hurst exponent, and variance ratio tests to validate stationarity. These checks confirm whether a price series is statistically likely to revert to a long-run average.
Mean reversion half-life is calculated using AR(1)-based estimation and Ornstein-Uhlenbeck parameter mapping. This determines the expected reversion horizon and feeds directly into z-score threshold logic for entry and exit signals.
Yes, Engle-Granger cointegration is used for pairs trading spread screening. A scanner evaluates pairwise combinations for correlation, spread Hurst, spread half-life, and z-score signals across multiple candidate assets.
Z-score entry and exit signals are generated by applying z-score threshold logic to validated time series. This transforms price or spread deviations into actionable long, short, or stop signals.
Yes, mean reversion analysis applies to crypto analytics. It supports single-asset crypto research and cointegrated pair spread screening across multiple candidate tokens using validated input time series.
0xZKnwGenerate trading signals from historical month-of-year and day-of-week effects.
mahmoud20138Identify mean-reversion opportunities in price data using Bollinger, RSI, and Z-score methods.
darkounus90Identify and analyze stock pairs using cointegration and correlation tests.
Superior-TradeAdd regime-based triple confirmation filters to directional trading strategies.
