How do you define an outlier in practice?
mediumAnswer
- No single definition — depends on the model and question.
- Common rules: (1) Tukey: outside Q1 - 1.5·IQR or Q3 + 1.5·IQR.
- (2) Z-score: |z| > 3 for approximately normal data.
- (3) Modified z-score using MAD: |z| > 3.5 (more robust).
- (4) Isolation Forest / DBSCAN for multivariate.
- Never remove outliers blindly — investigate whether they're data errors, natural extremes, or a signal.
Check yourself — multiple choice
- Anything > mean
- Tukey IQR rule / |z|>3 / MAD z-score / Isolation Forest — depends on data + model; investigate before removing
- Random threshold
- Always remove
Outlier: multiple rules (IQR, z, MAD, iForest); investigate before removing.
#descriptive#eda
Practise Statistics Fundamentals
215 interview questions in this topic.