GLM · 3D Distributions

Generalised Linear Models · 3D Posterior Landscape · Interactive · Rotate · Zoom

© Dr. Rainer Düsing · Interactive Tools by Claude

0.5
0.4
0.8
01 Bar Histogram
🖱 Rotate / Zoom
02 Curves / Surfaces
🖱 Rotate / Zoom
ℹ GLM 3D Visual — Help
What will I learn here?
Generalised Linear Models differ in their conditional distribution — not every outcome variable is normally distributed. This tool shows the 3D posterior landscape P(y | x, β₀, β₁) for six GLM families: How does the distribution shape change when the predictor x or the parameters vary?
Recommended exploration
The six GLM families
Normal (OLS): continuous data — symmetric bell curve, σ constant
Poisson: count data (0,1,2,…) — log link, Variance = Mean
Neg. Binomial: count data with overdispersion — Variance > Mean
Gamma: positive continuous data — right-skewed, log link
Binomial: number of successes out of n trials — logit link
Logistic: binary 0/1 data — probability between 0 and 1