StackedML
Practice
Labs
Questions
Models
Pricing
Sign in
Questions
/
Model Evaluation & Experimentation
/
Metrics (Regression)
/
R²
← Previous
Next →
533.
Negative R-Squared
easy
Can R² be negative? If so, when?
A
No — negative R² would imply negative variance which is mathematically undefined
B
Yes — but only when the model is trained on a different dataset than it is evaluated on
C
No — R² is bounded between 0 and 1 by the Cauchy-Schwarz inequality for all regression models
D
Yes — when the model performs worse than predicting the mean, SS_res exceeds SS_tot
Sign in to verify your answer
← Back to Questions