analyzing-time-series

What problem does it solve?

This Skill analyzes time-series CSV data to reveal its forecasting readiness by diagnosing stationarity, seasonality, trend, forecastability, and recommended transformations.

Core Features & Use Cases

• Data quality assessment: reports length, missing values, and detected frequency.
• Distribution characterization: mean, median, standard deviation, min/max, skewness, and kurtosis.
• Stationarity guidance: ADF/KPSS results and recommended differencing order.
• Seasonality detection: period estimation and strength assessment.
• Trend analysis: direction and strength of underlying trend.
• Autocorrelation insights: ACF/PACF patterns to inform model selection.
• Forecastability check: Ljung-Box based assessment of forecastability.
• Transform recommendations: Box-Cox analysis and variance-stability guidance.
• Use Case: Validate a CSV time-series before building ARIMA/SARIMA models and generate ready-to-use diagnostics.
• Use Case: QA data pipelines by quickly summarizing data quality and seasonal patterns.

Quick Start

• Run diagnostics on your CSV: python scripts/diagnose.py data.csv --output-dir results/
• Generate diagnostic plots: python scripts/visualize.py data.csv --output-dir results/
• Review outputs: open results/summary.txt and results/diagnostics.json for actionable insights