computeAccuracy: Get Measures of Forecast Accuracy

In mpascariu/MortalityForecast: Standard Tools to Compare and Evaluate Various Mortality Forecasting Methods

Description

Get Measures of Forecast Accuracy

Usage

computeAccuracy(u, u.hat, b, measures = NULL, xa = NULL, ya = NULL,
  na.rm = TRUE)

Arguments

u	Validation dataset.
u.hat	Out-sample forecast data.
b	Benchmark forecast data. Usually a naive or Random-Walk w drift forecast.
measures	What accuracy measure to compute? Various alternatives are available, Mean error measures: "ME", "MAE", "MAPE", "sMAPE", "sMRAE", "MASE"; Median error measures: "MdE", "MdAE", "MdAPE", "sMdAPE", "sMdRAE", "MdASE"; Squared error measures: "MSE", "RMSE", "RMSPE", "RMdSPE"; Geometric mean measure for positive errors: "GMRAE". If measures = NULL all the measures will be computed.
xa	Ages to be considered in model accuracy evaluation. It can be used to calculate the measures on a subset of the results. If xa = NULL (default) the entire age-range in input is considered.
ya	Years to be considered in accuracy computation. Default: ya = NULL.
na.rm	A logical value indicating whether NA values should be stripped before the computation proceeds.

Details

See \insertCitehyndman2006;textualMortalityForecast for a comprehensive discussion of the accuracy measures.

Author(s)

Marius D. Pascariu

References

\insertAllCited

mpascariu/MortalityForecast documentation built on Sept. 28, 2020, 2:40 p.m.