2. In general

g{E(yi)} = Xiβ where yi ∼ Exponential family 2. g is a known, smooth monotonic link function. 3. Xi is the ith row of a known model matrix, which depends on measured predictor variables (covariates). 4. β is an unknown parameter vector, estimated by MLE. 5. Xβ ( = η) is the linear predictor of the model. 6. Class includes log-linear models, general linear regression, logistic regression,... 7. glm in R implements this class. Generalized additive models, GAM 1. A GAM (semi-parametric GLM) is a GLM where the linear predictor depends linearly on unknown smooth functions