Fitting negative binomial

The negative binomial allows for the variance to exceed the mean, which is what you have measured in the previous exercise in your data crab. In this exercise you will recall the previous fit of the Poisson regression using the log link function and additionally fit negative binomial model also using the log link function.

You will analyze and see how the statistical measures were changed.

The model crab_pois and crab is loaded in your workspace.

This exercise is part of the course

Generalized Linear Models in Python

View Course

Exercise instructions

Define a formula for the regression model so that sat is predicted by width.
Fit the negative binomial using NegativeBinomial() and save the model as crab_NB.
Print the summaries of the Poisson model crab_pois and the newly fitted negative binomial model.

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# Define the formula for the model fit
formula = '____ ~ ____'

# Fit the GLM negative binomial model using log link function
crab_NB = smf.glm(formula = ____, data = ____, 
				  family = ____.____.____).____

# Print Poisson model's summary
print(____.____)

# Print the negative binomial model's summary
print(____.____)

Edit and Run Code