accuracy.gts: In-sample or out-of-sample accuracy measures for forecast grouped and hierarchical model

Description

Returns a range of summary measures of the forecast accuracy. The function measures out-of-sample forecast accuracy based on (holdout data - forecasts) and in-sample accuracy at the bottom level when setting keep.fitted = TRUE in the forecast.gts. All measures are defined and discussed in Hyndman and Koehler (2006).

Usage

# S3 method for gts
accuracy(object, test, levels, ..., f = NULL)

Arguments

object

An object of class gts, containing the forecasted hierarchical or grouped time series. In-sample accuracy at the bottom level returns when test is missing.

test

An object of class gts, containing the holdout hierarchical time series

levels

Return the specified level(s), when carrying out out-of-sample

...

Extra arguments to be ignored

Deprecated. Please use object instead.

Value

Matrix giving forecast accuracy measures.

Mean Error

RMSE

Root Mean Square Error

MAE

Mean Absolute Error

MAPE

Mean Absolute Percentage Error

MPE

Mean Percentage Error

MASE

Mean Absolute Scaled Error

Details

MASE calculation is scaled using MAE of in-sample naive forecasts for non-seasonal time series, and in-sample seasonal naive forecasts for seasonal time series.

References

R. J. Hyndman and A. Koehler (2006), Another look at measures of forecast accuracy, International Journal of Forecasting, 22, 679-688.

Examples

Run this code

# NOT RUN {
data <- window(htseg2, start = 1992, end = 2002)
test <- window(htseg2, start = 2003)
fcasts <- forecast(data, h = 5, method = "bu")
accuracy(fcasts, test)
accuracy(fcasts, test, levels = 1)

# }

Run the code above in your browser using DataCamp Workspace