Provides deterministic forecasting for weekly, monthly, quarterly, and yearly time series using the Generalized Adaptive Capped Estimator.
Includes a structured preprocessing pipeline with support for handling of non-positive or missing values, optional interpolation, and optional winsorization of extreme observations.
Implements multiple growth components including year-over-year, short-term movement, rolling-window behavior, and long-run drift.
Growth components are combined using a trimmed, robust averaging framework to ensure stable signal extraction across different series types.
Includes volatility-aware asymmetric caps that adapt to series characteristics and frequency.
Provides optional seasonal scaling using smoothed, normalized seasonal factors derived from the historical pattern.
Forecast generation uses a recursive formulation incorporating growth moderation (gamma) and level–growth blending (beta).
Includes a user-facing forecasting interface (gace_forecast) that returns forecasts in a consistent structure.
Includes a plotting helper (plot_gace) for visualizing historical and projected values using ggplot2.
Package includes documentation, examples, a vignette, and tests, and is structured for CRAN compatibility.