Poisson model

Description

Regression with Poisson model is a simple model used for count data.

Definition

For discrete positive outcome \(y\) and predictors \(x\), the model is:

\[ \begin{align} y_i &\sim \text{Poisson}(\exp{\eta_i}) \\ \eta_i &= \alpha + \beta \cdot x_i \end{align} \]

Parameters needing priors

  • \(\alpha\) (intercept)
  • \(\beta\) (predictor weights)

See also