Monte Carlo methods are important components of the data scientist's toolbox. Such techniques are used in risk analysis, decision making, and obviously pre-conditioning data for training a machine learning model.

In this chapter, you learned about generating normally distributed random values using the Box-Muller technique, applying Monte Carlo sampling to numerical integration, leveraging bootstrapped samples with replacements to construct statistics and finally extending sampling to a large dimension dataset using Markov Chain Monte Carlo.

This chapter completes the study of generative machine learning algorithms. The next part of the book introduces discriminative models, starting with the ubiquitous linear and logistic regression classifier.