Geoscientific model development· 2022Q1
Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not
- 1,795citations
- Q1SCImago
- 2022year
Short summary
Neither RMSE nor MAE is universally superior; RMSE is optimal for Gaussian errors, while MAE is optimal for Laplacian errors, with other metrics preferred for different error distributions.
AI-generated from the title and abstract; the full text is not read.
Key points
- RMSE and MAE are widely used but often confused model evaluation metrics.
- RMSE is optimal when model errors follow a Gaussian (normal) distribution.
- MAE is optimal when model errors follow a Laplacian distribution.
- Neither metric is inherently better than the other; their suitability depends on the error distribution.
AI-generated from the title and abstract; the full text is not read.
Abstract
Abstract. The root-mean-squared error (RMSE) and mean absolute error (MAE) are widely used metrics for evaluating models. Yet, there remains enduring confusion over their use, such that a standard practice is to present both, leaving it to the reader to decide which is more relevant. In a recent reprise to the 200-year debate over their use, Willmott and Matsuura (2005) and Chai and Draxler (2014) give arguments for favoring one metric or the other. However, this comparison can present a false dichotomy. Neither metric is inherently better: RMSE is optimal for normal (Gaussian) errors, and MAE is optimal for Laplacian errors. When errors deviate from these distributions, other metrics are superior.
The authors' abstract, as published at the source. Geoscientific model development, 2022 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Statistics and Probability
Statistics and ProbabilityMathematics