The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
Introduction to GACE: Generalized Adaptive Capped Estimator
Overview The Generalized Adaptive Capped Estimator (GACE) is a deterministic forecasting framework designed for stable and interpretable projections across weekly, monthly, quarterly, and yearly time series.
Unlike stochastic models, GACE uses a structured set of growth components and volatility-aware caps that ensure consistent behavior, even for noisy business or operational data.
This vignette introduces: • preprocessing and growth extraction, • growth capping, • seasonal factor construction, • recursive forecast generation, • how to use gace_forecast() and plot_gace().
Basic Usage library(GACE)
set.seed(1) y <- ts(rnorm(60, mean = 100, sd = 10), frequency = 12)
fc <- gace_forecast( df = y, periods = 12, freq = “month” )
head(fc)
Plotting the Forecast plot_gace(fc)
How GACE Works
Preprocessing GACE optionally cleans the input using: • winsorization of extreme values, • preservation of zeros, • minimal smoothing.
Growth Components GACE extracts four growth signals: • Year-over-year • Short-term (lag-1) • Rolling-window movement • Drift / long-run trend
These signals are trimmed, averaged, and dynamically stabilized.
Volatility-Aware Caps GACE applies asymmetric caps such as −0.30 to +0.30. Caps adapt when series volatility increases.
Seasonal Factors If seasonal = TRUE and frequency > 1, GACE computes smoothed, normalized seasonal loads.
Recursive Forecast future = level * (1 + growth_t) * seasonal_factor_t
with decay parameters gamma and beta.
Multi-frequency Support
Weekly → “week” Monthly → “month” Quarterly → “quarter” Yearly → “year”
Conclusion GACE provides a transparent, reproducible, and fast method for forecasting operational time series.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.