The Wonderful World of ML #3, Logistic Regression: Binary Classifier Workhorse
Details
Will the Broncos (Rockies or Avalanche) win or lose their next home game? Will Chris default on their credit card payment next month? These are just a couple examples of binary classification prediction problems. Logistic regression provides a straightforward way to model the probability that one of two mutually exclusive outcomes will occur. This is one of the simplest and most popular classifiers and serves as a good starting point for more sophisticated classification techniques such as LDA and QDA which we'll cover in the next session.
In this session we'll explore answers to the following question: How is a logistic regression model constructed? How do we find the best fit parameters to a logistc regression model? How do we handle discrete and continuous predictors in the model? What tools are available in R and Python to do logistic regression?
If any of you are still having trouble with the jupyter notebooks and would like some help, let me know and I'll schedule an informal session to do a hands-on session to make sure you get up and running.