StackedML
Practice
Labs
Questions
Models
Pricing
Sign in
Questions
/
Supervised Learning
/
Linear Models
/
Linear regression (OLS, assumptions)
← Previous
Next →
598.
Perfect Multicollinearity in OLS
medium
What happens to OLS coefficient estimates when two features are perfectly correlated?
A
The matrix XᵀX becomes singular and the OLS solution is undefined or non-unique
B
The coefficients are biased toward zero since the correlated features share predictive credit equally
C
The coefficients are estimated correctly but their standard errors become artificially small
D
The model automatically drops one of the correlated features to restore identifiability
Sign in to verify your answer
← Back to Questions