MathLiveMean absolute error
Gives the average size of the prediction errors, where each error is the absolute difference between a forecast f_i and the actual value y_i, so direction is ignored. It is used to measure how accurate a forecast or model is, in the same units as the data.
\mathrm{MAE} = \frac{1}{n}\sum_{i=1}^{n} \left| f_i - y_i \right| = \frac{1}{n}\sum_{i=1}^{n} \left| e_i \right|What the symbols mean
| number of observations or forecasts | |
| forecast or predicted value for observation i | |
| actual observed value for observation i | |
| forecast error for observation i, \(e_i = f_i - y_i\) |
About this formula
- FieldStatistics
- TopicForecast accuracy and error measures
- LevelUniversity
- Also calledMAE, mean absolute deviation of forecast errors, average absolute error, forecast error MAE formula, MAE formula, mean absolute error formula
- SourceWikidata Q6803609
Write it in Obsidian
MathLive is a free visual math editor for Obsidian. Run Find a formula, type “mean absolute error”, and it goes into your note as editable LaTeX. You can also scan formulas from screenshots, PDFs and handwriting.