MASE: MASE - Mean Absolute Scaled Error
In jdestefani/UEMTS: Univariate Error Measures for Time Series forecasting

Description Usage Arguments Value Examples

MASE = 1/T ∑_{t=1}^T ( ( e_t )/((T/(T-1)) ∑_{i=2}^T Y_i-Y_{i-1}))

1	MASE(y, y_hat, result_type = c("measure", "raw"))

`y`	True values of the time series
`y_hat`	Forecasted values of the time series
`result_type`	String indicating whether the function should return the raw values ("raw"), or the computed measure ("measure")

According to the value of result_type:

measure -> MASE computed according to the formula
raw -> Vector containing one MASE for each time step

y <- AirPassengers
splitting_point <- round(2*length(y)/3)
y_train <- y[1:splitting_point]
h <- 5

y_test <- y[(splitting_point+1):(splitting_point+h)]
y_hat_naive <- rep(tail(y_train,1),h)
results <- MASE(y_test,y_hat_naive)