MathLiveRoot-mean-square deviation
Gives the typical size of the difference between predicted values and observed values as a single number in the same units as the data. It is used to measure how accurate a forecast or model fit is, with smaller values meaning better agreement.
\mathrm{RMSD}=\sqrt{\frac{\sum_{t=1}^{T}(\hat{y}_t-y_t)^2}{T}}What the symbols mean
| root-mean-square deviation, in the same units as the data | |
| number of observations or time points | |
| index of the observation, running from 1 to T | |
| predicted or estimated value at time t | |
| observed (actual) value at time t |
About this formula
- FieldStatistics
- TopicForecasting and model evaluation
- LevelUniversity
- Also calledRMSE, root-mean-square error, RMSD formula, root mean square error formula, RMSE formula, forecast error measure
- SourceWikidata Q29037
Write it in Obsidian
MathLive is a free visual math editor for Obsidian. Run Find a formula, type “root-mean-square deviation”, and it goes into your note as editable LaTeX. You can also scan formulas from screenshots, PDFs and handwriting.