Katkat
Back

Diagnose Multicollinearity in Flight Delay Prediction Model

Capital OneData Scientist · Onsite
MediumData ScientistOnsiteMachine Learning

Last updated

Evaluates flight-delay modeling, data quality checks, and multicollinearity diagnosis. Strong answers identify leakage and timestamp issues, choose classification or regression based on the target, diagnose VIF and correlated features, and mitigate multicollinearity based on prediction or interpretation goals.

Scenario

You are asked to build a model that predicts whether a flight will be delayed using historical flight and weather data.

Question

Inspect the raw dataset and list any data-quality issues you notice (e.g., missing values, impossible seat counts, weekday encoded as numeric). Choose an appropriate modeling framework and justify classification versus regression for the stated outcome. VIF scores show high multicollinearity; describe how you would diagnose and mitigate this problem when presenting to another data scientist. In an ideal setting you can run an experiment—outline the experimental design that would help solve or confirm the multicollinearity issue.

Hints

Mention imputation, data validation, one-hot encoding, feature selection, regularization, variance inflation factors, and A/B or switchback tests.