mase: MASE
In max-graham/metrics: A Collection of Useful Metrics

Description Usage Arguments Value Note Examples

Calculates the Mean Absolute Scaled Error(MASE) between x and the target. This is intended to be used for measuring the accuracy of a time series forecast, since it reports the error in the context of the naive forecast (random walk for non-seasonal, or a seasonal naive otherwise).

1	mase(x, target, m = 1, na.rm = FALSE)

`x`	numeric vector; the forecast to evaluate.
`target`	numeric vector; the target values.
`m`	atomic integer; the seasonal period. Defaults to 1 if non-seasonal.
`na.rm`	logical; should NAs be removed before calculation?

An atomic numeric vector containing the calculated MASE.

This will return Inf when the last length(x) - m values in target are all the same.

## Not run: mase(c(1, 2, 3))            #> Error: different lengths (3, 0)
mase(c(1, 2, 3), c(1, 4, 4))          #> 0.66667
mase(c(1, 2, NA, 4), c(2, 2, 5, NA))  #> NA
mase(c(1, 2, 3, NA, 4),
     c(2, 2, 5, NA, NA),
     na.rm = TRUE)                    #> 0.66667