Summary
This chapter barely scratches the surface of the topic of generalized linear models with the description of linear and logistic regression algorithms. Regression models, along with Naïve Bayes classification, are the most well-understood techniques by those without a deep knowledge of statistics or machine learning.
At the end of this chapter, you hopefully have a grasp of the following:
Linear and non-linear least squares-based optimization
The implementation of ordinary least square regression, as well as logistic regression as classifiers and predictive models
The purpose of regularization as illustrated with ridge regression
The regression models do not impose the condition that the features have to be independent, contrary to the Naïve Bayes models (refer to Chapter 6, Naïve Bayes Classifiers). However, these models do not take into account the sequential nature of time series commonly used in dynamic asset pricing. The next chapter introduces models for sequential data, with two...
