Advantages of the MAD/Mean ratio over the MAPE

By Stephan Kolassa, W. Schütz. 2007

101 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/e56d2fec69b80672f36b57e85f59b2b86404a786

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. S. Prestwich
2. Robert Fildes
3. Shaohui Ma
4. Roberto Rossi
5. Stephan Kolassa
6. S. Armagan Tarim
7. Brahim Hnich
8. Rob J Hyndman

Citations in the corpus

• Mean-based error measures for intermittent demand forecasting (2013, methodology): “However, Kolassa and Schütz (2007) note that the MASE of two series with identical forecasts and identical demands during the forecast horizon will differ if the two series differed in their historical demands.”
• Performance Metrics (Error Measures) in Machine Learning Regression, Forecasting and Prognostics: Properties and Typology (2018, unknown): “- MAD/Mean ratio (Hoover, 2006; Kolassa and Schütz, 2007).”
• Retail forecasting: Research and practice (2019, background): “But as Kolassa (2017) has pointed out, the technical hurdles of translating individual (as opposed to segmented) demand histories into aggregate SKU level forecasts have yet to be successfully overcome.”