mase {lfl}

Compute Mean Absolute Scaled Error (MASE)

Description

MASE is computed as sum(abs(validation - forecast)) / sum(abs(validation[-1] - validation[-n])) / (n/(n-1)).

Usage

mase(forecast, validation)

Arguments

forecast

A numeric vector of forecasted values

validation

A numeric vector of actual (real) values

Value

A Mean Absolute Scaled Error (MASE)

Author(s)

Michal Burda

See Also

rmse(), smape(), frbe()