Mean absolute percentage error

The mean absolute percentage error (MAPE), also known as mean absolute percentage deviation (MAPD), is a measure of prediction accuracy of a forecasting method in statistics, for example in trend estimation. It usually expresses accuracy as a percentage, and is defined by the formula:

\mbox{M} = \frac{1}{n}\sum_{t=1}^n  \left|\frac{A_t-F_t}{A_t}\right|,

where At is the actual value and Ft is the forecast value.

The difference between At and Ft is divided by the Actual value At again. The absolute value in this calculation is summed for every forecasted point in time and divided by the number of fitted points n. Multiplying by 100 makes it a percentage error.

Although the concept of MAPE sounds very simple and convincing, it has major drawbacks in practical application [1]

Alternative MAPE definitions

Problems can occur when calculating the MAPE value with a series of small denominators. A singularity problem of the form 'one divided by zero' and/or the creation of very large changes in the Absolute Percentage Error, caused by a small deviation in error, can occur.

The difference with the original formula is that each Actual Value (At) of the series is replaced by the average Actual Value (Āt) of that series. Hence, the distortions are smoothed out. This alternative is still being used for measuring the performance of models that forecast spot electricity prices.[2]

Another variant is the sum of absolute differences is divided by sum of actual values. Sometimes, it is referred as WAPE.

Issues

While MAPE is one of the most popular measures for forecasting error, there are many studies on shortcomings and misleading results from MAPE.[3] First the measure is not defined when the actual value is zero, A_t=0. Moreover, MAPE puts a heavier penalty on negative errors, A_t < F_t than on positive errors.[4] To overcome these issues with MAPE, there are some other measures proposed in literature:

See also

External links

References

  1. 1 2 Tofallis (2015). "A Better Measure of Relative Prediction Accuracy for Model Selection and Model Estimation", Journal of the Operational Research Society, 66(8),1352-1362. archived preprint
  2. Jorrit Vander Mynsbrugge (2010). "Bidding Strategies Using Price Based Unit Commitment in a Deregulated Power Market", K.U.Leuven
  3. Hyndman, Rob J., and Anne B. Koehler. "Another look at measures of forecast accuracy." International journal of forecasting 22.4 (2006): 679-688.
  4. Makridakis, Spyros. "Accuracy measures: theoretical and practical concerns." International Journal of Forecasting 9.4 (1993): 527-529
This article is issued from Wikipedia - version of the Saturday, April 09, 2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.