QDA vs LDA — how do you choose between them?
mediumAnswer
- QDA relaxes LDA's shared-covariance assumption: each class has its own , so the decision boundary is quadratic.
- Prefer QDA when class covariances clearly differ (visible from ellipsoid plots or per-class covariance estimates) and you have enough data per class to estimate them reliably.
- Prefer LDA when data per class is limited, features are noisy, or class covariances are similar — the pooled Σ regularizes the estimate.
Check yourself — multiple choice
- QDA is always better
- QDA fits per-class covariances (quadratic boundaries); prefer LDA when data per class is small
- LDA is quadratic
- They are identical
QDA = per-class covariance ⇒ quadratic boundary; LDA is more data-efficient with pooled Σ.
#lda-qda#discriminant-analysis
Practise Supervised Learning
215 interview questions in this topic.