R-Squared (R²)
Definition
R² describes the variation of a target variable explained in the model context.
A metric that expresses the proportion of variation explained by a regression model in the respective model context.
Simply explained
The metric shows how much of the observed differences the regression model explains compared to its baseline measure.
Why it matters in practice
The term helps to select, develop, evaluate AI systems in a well-founded manner and realistically assess their technical risks.
Example from business
A model explains 72 percent of the variation of a target variable in the test set.
Limits and pitfalls
- Interpreting R² as causality or universal quality, using training data, or comparing variants without validation.
Origin & history
R² has been widespread since early regression analysis and has been further developed in many model variants.
Use in mid-sized companies
- Possible application: A model explains 72 percent of the variation of a target variable in the test set.
Frequently asked questions
What does coefficient of determination (R²) mean?
A metric that expresses the proportion of variation explained by a regression model in the respective model context. What mistakes should be avoided? Interpreting R² as causality or universal quality, using training data, or comparing variants without validation.
Sources
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: