On the asymmetry of the symmetric MAPE

By Paul Goodwin, R. Lawton. 1999

Subfields: Statistics and Probability, Information Theory

How researchers see this paper

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

In Statistics and Probability

In Information Theory

Citations in the corpus

Forecasting with a hybrid method utilizing data smoothing, a variation of the Theta method and shrinkage of seasonal factors (2018, methodology): “The Mean Absolute Scaled Error (MASE) proposed by Hyndman and Koehler (2006) was also used for comparing the accuracy of forecasting methods such as sMAPE, which has been proven to penalize large positive errors more than negative ones and to increase the difference of their absolute size (Goodwin a”
The accuracy of intermittent demand estimates (2005, background): “The symmetric MAPE could be calculated, but it is known to suffer from asymmetry problems ( Goodwin & Lawton, 1999 ).”
Advances in forecasting with neural networks? Empirical evidence from the NN3 competition on time series prediction (2011, methodology): “Despite the shortcomings of the sMAPE (Goodwin & Lawton, 1999), it was chosen both because it served as the primary criterion in the M3 competition and to make the NN3 results accessible to practitioners, whose predominant error metric is the MAPE.”
Another look at measures of forecast accuracy (2006, background): “See Goodwin and Lawton (1999) and Koehler (2001) for further discussion on this point.”
25 Years of IIF Time Series Forecasting: A Selective Review (2005, background): “See Goodwin & Lawton (1999) and Koehler (2001) for further discussion on this point.”
25 years of time series forecasting (2006, background): “See Goodwin & Lawton (1999) and Koehler (2001) for further discussion on this point.”
Performance Metrics (Error Measures) in Machine Learning Regression, Forecasting and Prognostics: Properties and Typology (2018, background): “However, later it was shown that the objective was not gained – sMAPE (symmetric MAPE) was still asymmetric (Goodwin and Lawton, 1999).”
Theta intelligent forecasting information system (2003, methodology): “SMAPE (Makridakis and Hibon, 2000; Goodwin and Lawton, 1999) is computed for each forecasting horizon, and that is the extrapolation error level estimation of the model.”
To combine or not to combine: selecting among forecasts and their combinations (2005, methodology): “Strengths and weaknesses of this accuracy measure have been reported and criticized before, for example by Goodwin and Lawton (1999) and Koehler (2001).”
The use of analogies in forecasting the annual sales of new electronics products (2013, background): “However, the widely used mean absolute percentage error (MAPE) is distorted when some periods have very low sales (e.g. see Goodwin and Lawton, 1999).”