takes an input vector 𝑥∊ℝ^𝑛 as input and predicts the value of a count 𝑦∊ℝ as output
a type of parametric regression model whose dependent variable is a count of items, events, results, or activities
counts are nonnegative integers (0, 1, 2, etc.)
count data with:
- higher means tend to be normally distributed and you can often use Linear Regression/OLS
- smaller means can be skewed, and Linear Regression might have a hard time fitting these data. In these cases, we use count regression models

Count Models - Types

Count Model Type	Description
Poisson Regression Model	count data frequently follow the Poisson distribution, which makes Poisson Regression a good possibility
Negative Binomial Regression Model (NB2)	used when there is over-dispersion (i.e. mean and variance are different) Poisson Regression assumes that the variance equals the mean. When the variance is greater than the mean, your model has over-dispersion. NB2 can be more appropriate when over-dispersion is present
Zero-Inflated Models	used when there is a lot of 0 counts count data might have too many zeros to follow the Poisson distribution. Zero-Inflated Models can be more appropriate in modeling this

Resources