Apply offsets and weights as appropriate.

Free SOA Exam PA (Predictive Analytics) lesson in Generalized Linear Models. 16 min read, ~2,463 words.

An offset is a predictor whose coefficient is fixed at 1, added on the link scale, never estimated. Use a log offset (log of exposure) in a Poisson count model to model a rate per unit of exposure. A weight scales how much each observation contributes to the likelihood; it...

Read the full lesson, free →
Worked examples, audio narration, and practice. Free with a free account, no card.

What this lesson covers

Learning objectives

Browse all free Exam PA lessons or jump into free Exam PA practice questions.