## Contents

2. Linear Models for Continuous Data

2.2. Estimation of the Parameters

2.3. Tests of Hypotheses

2.4. Simple Linear Regression

2.5. Multiple Linear Regression

2.6. One-Way Analysis of Variance

2.7. Two-Way Analysis of Variance

2.8. Analysis of Covariance Models

2.9. Regression Diagnostics

2.10. Transforming the Data

3. Logit Models for Binary Data

3.2. Estimation and Hypothesis Testing

3.3. The Comparison of Two Groups

3.4. The Comparison of Several Groups

3.5. Models With Two Predictors

3.6. Multi-factor Models: Model Selection

3.7. Other Choices of Link

3.8. Regression Diagnostics for Binary Data

4. Poisson Models for Count Data

4.2. Estimation and Testing

4.3. A Model for Heteroscedastic Counts

5. Log-Linear Models for Contingency Tables

6. Multinomial Response Models

6.2. The Multinomial Logit Model

6.3. The Conditional Logit Model

6.4. The Hierarchical Logit Model

6.5. Models for Ordinal Response Data

7.2. Censoring and The Likelihood Function

7.3. Approaches to Survival Modeling

7.4. The Piece-Wise Exponential Model

7.5. Infant and Child Mortality in Colombia

7.6. Discrete Time Models

## Browsing or Printing?

The notes are offered in two formats: HTML and PDF, see the discussion below for more details.

I expect most of you will want to *print* the notes, in which case
you can use the links below to access the PDF file for each chapter:

If you are *browsing* you can use the table of contents on the right
to go directly to a specific chapter or section.

No, there is no Chapter 1 ... yet. One day I will write an introduction to the course and that will be Chapter 1.

* The list above has two extensions to the original notes:
an addendum on
*Modeling Over-Dispersed Count Data*, which describes models with
extra-Poisson variation and negative binomial regression,
and a brief discussion
of models for longitudinal and clustered data.

## Suggested Citation

RodrÃguez, G. (2007). *Lecture Notes on Generalized Linear Models*.
Available at http://data.princeton.edu/wws509/notes/

## The Choice of Formats

It turns out that making the lecture notes available on the web was a bit of a challenge because web browsers were designed to render text and graphs but not equations, which are often shown using bulky graphs or translated into text with less than ideal results.

The notes were written using LaTeX, which produces postscript or PDF,
so the simplest solution was to post the generated PDF files, one per
chapter. This format is best for *printing* the notes.
Our PDF files are now smaller and look better on the screen that before.
To view the files you will need
Adobe Reader,
unless you use a browser like Chrome that can render PDF natively.

I also wanted to offer a version in HTML, which is best for *browsing*.
Fortunately, there is now an excellent Javascript library called
MathJax
that does a superb job of rendering mathematics on the web.
I combined this library with a custom program that translates the rest of
the LaTeX source into HTML, leaving all equations to MathJax.
This approach supersedes the original pages, which had been generated using
TtH, a program that
worked much better than alternatives such as LaTeX2Html and served us
us well for many years.

Continue with 2. Linear Models for Continuous Data