6.10 Summary
In this chapter, we have seen how to use Bambi to fit Bayesian models as an alternative to the pure PyMC model. We start with the simplest case, a model with a single predictor, and then move to more complex models, including polynomials, splines, distributional models, models with categorical predictors, and interactions.
The main advantage of Bambi is that it is very easy to use; it is very similar to R’s formula syntax. And internally, Bambi defines weakly informative priors and handles details that can be cumbersome for complex models. The main disadvantage is that it is not as flexible as PyMC. The range of models that Bambi can handle is a small subset of those from PyMC. Still, this subset contains many of the most commonly used statistical models in both industry and academia. The strength of Bambi is not just easy model building, but easier model interpretation. Across the chapter, we have seen how to use Bambi’s interpret module to gain a better understanding...
