Regression
Describes the relation between a response Y and one or several predictors X as a mathematical model — not just "do they move together" but a predictive equation.
Also known as: regression analysis, regression model
Regression yields a function estimated from observed data. It answers two questions: (1) prediction — what Y is expected at new X? (2) understanding — which X influence Y and by how much?
- Classical: linear regression (OLS) — continuous, normally distributed Y
- Attributive: Generalized Linear Models — binary, categorical or count data
- Non-linear — cubic splines, polynomial regression, non-linear regression
See also
Used in
In the Algorithm Lab
Sources
- Montgomery, D. C., Peck, E. A. & Vining, G. G. — Introduction to Linear Regression Analysis, 6th Edition, Wiley
- Fahrmeir, L., Kneib, T., Lang, S. & Marx, B. — Regression: Models, Methods and Applications, 2nd Edition, Springer