In this chapter, you got a brief overview of modern Bayesian machine learning and its applications in finance. We've only touched upon this as it is a very active field of research from which we can expect many breakthroughs in the near future. It will be exciting to observe its development and bring its applications into production.
Looking back at this chapter, we should feel confident in understanding the following:
- The empirical derivation of Bayes formula
- How and why the Markov Chain Monte Carlo works
- How to use PyMC3 for Bayesian inference and probabilistic programming
- How these methods get applied in stochastic volatility models
Notice how everything you have learned here transfers to bigger models as well, such as the deep neural networks that we've discussed throughout the entirety of the book. The sampling process is still a bit slow for very large models, but researchers are actively working on making it faster, and what you've learned is a great foundation for the future.