ABSTRACT
One of the most straightforward definitions of intermittent demand is that it is the demand that happens at irregular frequency. This chapter is based on Svetunkov and Boylan. There are two dimensions for the model selection in the oETS model: selection of the type of occurrence and selection of the model type for the occurrence part. The main difficulty with pure multiplicative models arises from the construction point of view. ADAM ARIMA can also be used for demand sizes, resulting in the iARIMA model. Finally, when it comes to using an intermittent state space model on count data, there is a temptation to round up the resulting forecasts. While there are different ways of estimating the parameters of the ADAM, it is worth focusing on likelihood estimation for consistency with other ADAMs. As a result, several challenges are related to the ADAM specifically and to the intermittent demand at large are worth discussing.
