Written in the form:
n - natural/canonical parameter
T(y) - sufficient statistic (usually = y)
a(n) - log partition function
Generalized Linear Models
Constructing GLMs
Assumptions
1) y | x; θ ~ ExponentialFamily(n).
2) Given x, our goal is to predict the expected value of T(y) given x.
3) The natural parameter n and the inputs x are related linearly
2) Given x, our goal is to predict the expected value of T(y) given x.
3) The natural parameter n and the inputs x are related linearly