mase: Mean Absolute Scaled Error (MASE)

In thfuchs/tsRNN: Time Series Forecasting using Recurrent Neural Networks by Keras

Description

Mean Absolute Scaled Error (MASE), scaled according to M4 Forecasting competition

Usage

mase(data, actual, forecast, m)

Arguments

data	time series (only train set)
actual	actual values (Equivalent of forecast)
forecast	predicted values
m	frequency, e.g. 12 for monthly and 4 for quarterly series

Value

numeric vector of length 1

References

• Svetunkov, I., Sagaert, Y. R. (2020). greybox: Toolbox for Model Building and Forecasting. https://cran.r-project.org/package=greybox

• Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2020). The M4 Competition: 100,000 time series and 61 forecasting methods. International Journal of Forecasting, 36(1), 54–74. https://doi.org/10.1016/j.ijforecast.2019.04.014

See Also

Other PF accuracy measures: mape(), smape()

Examples

data <- tsRNN::fc_arima[key == "actual", value]
forecast <- tsRNN::fc_arima[key == "predict", value]
data_train <- data[1:(length(data) - length(forecast))]
data_test <- data[(length(data) - length(forecast) + 1):length(data)]

mase(data = data_train, actual = data_test, forecast = forecast, m = 4)

thfuchs/tsRNN documentation built on April 17, 2021, 11:03 p.m.