Chapter 9: Statistics for Classification
In the previous chapter, we covered regression problems where correlations, in the form of a numerical relationship between independent variables and dependent variables, are established.
Different from regression problems, classification problems aim to predict the categorical dependent variable from independent variables. For example, with the same Netflix stock price data and other potential data, we can build a model to use historical data that predicts whether the stock price will rise or fall after a fixed amount of time. In this case, the dependent variable is binary: rise or fall (let's ignore the possibility of having the same value for simplicity). Therefore, this is a typical binary classification problem. We will look at similar problems in this chapter.
In this chapter, we will cover the following topics:
- Understanding how a logistic regression classifier works
- Learning how to evaluate the performance of...
