Forecast time series data with ARIMA, Prophet, and regression models.
npx skills add https://github.com/Phife726/ds_agent --skill ds-time-series
Please help me install this Agent Skill. Skill: ds-time-series Source: https://github.com/Phife726/ds_agent/tree/main/ds-time-series Command: npx skills add https://github.com/Phife726/ds_agent --skill ds-time-series
Guides users through the full forecasting lifecycle — from understanding temporal patterns to deploying production-grade forecast models.
Load your time-stamped data, perform initial EDA and stationarity checks, fit a chosen model, and generate forecasts with prediction intervals.
High-intent search queries and answers about installing and using this skill.
Time series forecasting with seasonality and trends uses models like ARIMA, SARIMA, ETS, and Prophet to capture temporal patterns. This Skill guides you through EDA, stationarity checks, and model fitting to generate predictive forecasts with uncertainty intervals.
Evaluating time series forecasting models requires time-aware train/test splits and metrics like MAE, RMSE, MAPE, and MASE. This approach prevents data leakage and accurately assesses predictive accuracy and interval quality on unseen temporal data.
Building a production-ready time series forecasting pipeline involves applying robust diagnostic checks and structured workflows across ETS, ARIMA, or Prophet models. This ensures stable deployment by automating EDA, fitting, and forecast generation steps.
Choosing between Prophet and ARIMA for time series forecasting depends on your data characteristics. This Skill supports model selection across ETS, ARIMA, SARIMA, Prophet, and ML approaches with lag features, allowing comparison to find the best fit.
Yes, you can use machine learning for time series prediction by engineering lag features and time-based regressors. This Skill supports ML approaches alongside traditional methods, enabling regression models to capture complex temporal dependencies.
Yes, checking stationarity before fitting ARIMA models is essential for valid results. This Skill includes initial EDA and stationarity checks as foundational steps, ensuring your timestamped data is properly prepared before model selection.
coolsocketAnalyze datasets with pandas, polars, seaborn, and matplotlib for cleaning, EDA, visualization, and modeling.
zachmayerForecast numeric time series data with R's auto.arima algorithm.
xjtulycBuild end-to-end time series analysis workflows for forecasting and model comparison.
ntluong95Specify and fit Distributed Lag Non-Linear Models for time series regression.
