CHAPTER 19 Other Useful Kinds of Regression 271

Chapter 19

Other Useful Kinds

of Regression

T

his chapter covers regression approaches you’re likely to encounter in bio-

statistical work that are not covered in other chapters. They’re not quite as

common as straight-line regression, multiple regression, and logistic regres-

sion (described in Chapters 16, 17, and 18, respectively), but you should be aware of

them. We don’t go into a lot of detail, but we describe what they are, the circum-

stances under which you may want to use them, how to execute the models and

interpret the output, and special situations you may encounter with these models.

Note: We also don’t cover survival regression in this chapter, even though it’s one

of the most important kinds of regression analysis in biostatistics. Survival analy-

sis is the theme of Part 6 of this book, and is the topic of Chapter 23.

Analyzing Counts and Rates

with Poisson Regression

Statisticians often have to analyze outcomes consisting of the number of occur-

rences of an event over some interval of time, such as the number of fatal highway

accidents in a city in a year. If the occurrences seem to be getting more common

IN THIS CHAPTER»

» ^{Using Poisson regression to analyze}

counts and event rates»

» ^{Getting a grip on nonlinear regression»}

» ^{Smoothing data without making any}

assumptions