Summary
In this chapter, we learned about probabilistic programming and how inference engines leverage the power of Bayesian modeling. We discussed the main conceptual ideas behind MCMC methods and its central role in modern Bayesian data analysis. We encountered, for the first time, the powerful and easy-to-use PyMC3 library. We revisited the coin-flipping problem from the previous chapter, this time using PyMC3 to define it, solve it, and also perform model checks and diagnoses that are a very important part of the modeling process.
In the next chapter, we will keep building our Bayesian analytics skills by learning how to work with models having more than one parameter and how to make parameters talk to each other.
