MEPX
Chapter 8 of 10All chapters
  1. 1Samples and populations
  2. 2The central limit theorem
  3. 3Confidence intervals
  4. 4Hypothesis testing
  5. 5Multiple comparisons
  6. 6A/B testing
  7. 7Causal inference
  8. 8Regression
  9. 9A Bayesian view
  10. 10Reporting results

Chapter 8 of 10

Regression

Modelling relationships.

What a coefficient means

The expected change in the outcome per unit change in that predictor, holding the others fixed. That last clause carries most of the difficulty.

  • Correlated predictors make individual coefficients unstable.
  • Extrapolating beyond the range of the data is not supported by it.

Diagnostics

Check residuals for pattern, look for influential points, and plot the data. A good fit statistic on a badly specified model is worse than no model.