Skip to contents

Croston with demand classification

Source: R/forecast_native.R
morie_joseph_croston_intermittent.Rd

morie_croston with the Syntetos-Boylan-Croston quadrant: average interval against squared coefficient of variation of the nonzero demands, split at 1.32 and 0.49.

Usage

morie_joseph_croston_intermittent(
  y,
  alpha = 0.1,
  variant = c("sba", "croston")
)

Arguments

y

non-negative demand series.

alpha

smoothing parameter in (0, 1].

variant

"croston" or "sba".

Value

list as morie_croston plus classification (smooth / erratic / intermittent / lumpy), cv_squared, average_interval.

Details

The classification is the useful part. Croston's method is designed for INTERMITTENT demand, and applying it to a smooth series makes things worse rather than better – ordinary exponential smoothing wins there, and the function says so.

References

Syntetos, A. A., Boylan, J. E. and Croston, J. D. (2005). On the categorization of demand patterns. JORS, 56(5), 495-503.

Examples

morie_joseph_croston_intermittent(c(0, 0, 5, 0, 0, 0, 3, 0, 4))$classification
#> [1] "intermittent"