You are here

Statistical Models: Theory and Practice

Cambridge University Press
Number of Pages: 

This is a revised edition of a book first published in 2005. See our review of that edition. According to Freedman's preface, "the exposition has been improved in a number of ways, without (I hope) introducing new difficulties. There are many new examples and exercises." In addition, some of the material has been reorganized; a look at the table of contents shows that these changes are minor.

What the table of contents does not show, however, is that solutions to the problems and the reprints of social science papers fill up half the book. The goal is definitely to have the intended readers (upper-level undergraduates and graduate students in statistics) interact directly with the data analysis in the examples.

One distinguishing feature of this book is the choice of examples: many of them deal with fairly controversial socio-political issues, presumably in the hope of engaging the students. Thus, the author discusses a study of which segments of society supported political repression in the McCarthy era, another suggesting that Catholic schools are much more effective than public schools, and so on. Instructors who are leery of getting into political discussions should probably pick another book!

Fernando Q. Gouvêa is Carter Professor of Mathematics at Colby College in Waterville, ME.

Date Received: 
Tuesday, May 26, 2009
Include In BLL Rating: 
David A. Freedman
Publication Date: 
Fernando Q. Gouvêa

Foreword to the Revised Edition


1. Observational studies and experiments
1.1 Introduction
1.2 The HIP trial
1.3 Snow on cholera
1.4 Yule on the causes of poverty
     Exercise set A
1.5 End notes

2. The regression line
2.1 Introduction
2.2 The regression line
2.3 Hooke's law
     Exercise set A
2.4 Complexities
2.5 Simple vs multiple regression
     Exercise set B
2.6 End notes

3. Matrix algebra
3.1 Introduction
     Exercise set A
3.2 Determinants and inverses
     Exercise set B
3.3 Random vectors
     Exercise set C
3.4 Positive definite matrices
     Exercise set D
3.5 The normal distribution
     Exercise set E
3.6 If you want a book on matrix algebra

4. Multiple regression
4.1 Introduction
     Exercise set A
4.2 Standard errors
     Things we don't need
     Exercise set B
4.3 Explained variance in multiple regression
     Association or causation?
     Exercise set C
4.4 What happens to OLS if the assumptions break down?
4.5 Discussion questions
4.6 End notes

5. Multiple regression: special topics
5.1 Introduction
5.2 OLS is BLUE
     Exercise set A
5.3 Generalized least squares
     Exercise set B
5.4 Examples on GLS
     Exercise set C
5.5 What happens to GLS if the assumptions break down?
5.6 Normal theory
     Statistical significance
     Exercise set D
5.7 The F-test
     "The" F-test in applied work
     Exercise set E
5.8 Data snooping
     Exercise set F
5.9 Discussion questions
5.10 End notes

6. Path models
6.1 Stratification
     Exercise set A
6.2 Hooke's law revisited
     Exercise set B
6.3 Political repression during the McCarthy era
     Exercise set C
6.4 Inferring causation by regression
     Exercise set D
6.5 Response schedules for path diagrams
     Selection vs intervention
     Structural equations and stable parameters
     Ambiguity in notation
     Exercise set E

6.6 Dummy variables
     Types of variables
6.7 Discussion questions
6.8 End notes

7. Maximum likelihood
7.1 Introdcution
     Exercise set A
7.2 Probit models
     Why not regression?
     The latent-variable formulation
     Exercise set B
     Identification vs estimation
     What if the Ui are N(μ,σ2)?
     Exercise set C
7.3 Logit models
     Exercise set D
7.4 The effect of Catholic schools
     Latent variables
     Response schedules
     The second equation
     Mechanics: bivariate probit
     Why a model rather than a cross-tab?
     More on table 3 in Evans and Schwab
     More on the second equation
     Exercise set E
7.5 Discussion questions
7.6 End notes

8. The bootstrap
8.1 Introduction
     Exercise set A
8.2 Bootstrapping a model for energy demand
     Exercise set B
8.3 End notes

9. Simultaneous equations
9.1 Introduction
     Exercise set A
9.2 Instrumental variables
     Exercise set B
9.3 Estimating the butter model
     Exercise set C
9.4 What are the two stages?
     Invariance assumptions
9.5 A social-science example: education and fertility
     More on Rindfuss et al
9.6 Covariates
9.7 Linear probability models
     The assumptions
     The questions
     Exercise set D
9.8 More on IVLS
     Some technical issues
     Exercise set E
     Simulations to illustrate IVLS
9.9 Discussion questions
9.10 End notes

10. Issues in statistical modeling
10.1 Introduction
     The bootstrap
     The role of asymptotics
     Philosophers' stones
     The modelers' response
10.2 Critical literature
10.3 Response schedules
10.4 Evaluating the models in chapters 7–9
10.5 Summing up


Answers to Exercises

The Computer Labs

Appendix: Sample MATLAB Code

     Gibson and McCarthy
     Evans and Schwab on Catholic Schools
     Rindfuss et al on Education and Fertility
     Schneider et al on Social Capital


Publish Book: 
Modify Date: 
Tuesday, May 26, 2009