Session 4: What Can Go Wrong — Biases and Diagnostics
Note
Date: Thursday, 07 May 2026, 16:00–19:00
Business question
Does the hotel price–distance relationship hold across all European cities, or does it break down in some?
Learning goals
- Identify key OLS assumption violations in practice
- Detect multicollinearity, heteroskedasticity, and influential outliers
- Understand omitted variable bias intuitively
- Apply robust standard errors using
lmtestandsandwich
Dataset
hotels-europe — 148,021 hotel-date observations across 46 European cities and 10 dates. The larger and more heterogeneous sample makes OLS assumption violations more visible.
Quarto skill introduced this session
Embedding diagnostic plots in a report: figure captions, cross-references (@fig-), and layout options for multi-panel figures.
Important
Material will be available here before the session.