Evaluation & Quality
Root mean squared error
A prediction-error score that gives extra weight to large mistakes and uses the original measurement units.
Example
A sales forecast's error score is reported in units sold.
Why people use it
It summarizes prediction errors while giving large misses extra influence.
What you'll hear
“Those occasional big misses really affect the score.”
What this means for you
Compare typical mistakes with the effect of occasional large misses.
Can you control it?
No
No direct control. This describes a wider issue, concept or result rather than something you can simply switch on or off in a tool.
Common questions
- Why does RMSE react strongly to large errors?
- Squaring the errors gives larger misses more influence before the square root is taken.
- Is the result in the original measurement units?
- Yes. For a forecast measured in dollars, RMSE is also expressed in dollars.
- Can it hide whether forecasts are too high or too low?
- Yes. Squaring errors removes their direction before the score is calculated.