Statistical Analysis with Excel For Dummies, 5th Edition

Chapter 21

Estimating Probability: Logistic Regression

IN THIS CHAPTER

Defining logistic regression

Checking out the XLMiner Analysis ToolPak

Working with logistic regression

In this chapter, I explore a type of regression that’s different from any regression analysis in Chapter 14. The regression you’ve already heard about involves a continuous dependent variable whose value you predict (in a linear way) from a continuous independent variable (or from a set of independent variables). You make that prediction on the basis of data on each independent variable.

In this new and different type of regression, the dependent variable is the probability of a “success” of a binary event — like, say, if a person decides to buy a product (success) or not (failure) after spending some time looking at an ad for the product. “Time spent looking at the ad” is the independent variable.

The goal is to estimate the probability of buying the product based on how much time the person looks at the ad. The dependent variable is continuous, but based on what you’ve learned about probability, you know it has a minimum value of 0 and a maximum value of 1.

This is called logistic regression, and as you’ll ...