Practical Multivariate Analysis, Fifth Edition
Authors: 
Abdelmonem Afifi, Susanne May, and Virginia A. Clark 
Publisher: 
Chapman & Hall/CRC 
Copyright: 
2012 
ISBN13: 
9781439816806 
Pages: 
517; hardcover 
Price: 
$72.75 



Comment from the Stata technical group
The fifth edition of Practical Multivariate Analysis, by Afifi,
May, and Clark, provides an applied introduction to the analysis of
multivariate data. The preface says:
“We wrote this book for investigators, specifically behavioral
scientists, biomedical scientists, and industrial or academic
researchers, who wish to perform multivariate statistical analyses and
understand the results. We expect readers to be able to perform and
understand the results, but also expect them to know when to ask for help
from an expert on the subject. It can either be used as a selfguided
textbook or as a text in an applied course in multivariate analysis.
Sections 1 and 2, the first half of the book, review the
basics: understanding the different types of data, preparing your data,
selecting appropriate statistical techniques, and using and understanding
regression and correlation techniques.
Section 3, the second half of the book, covers canonical correlation,
discriminant analysis, logistic regression, survival analysis, principal
components, factor analysis, cluster analysis, loglinear analysis,
and correlated outcomes regression (think xtmixed in Stata).
The applied introductory nature of the book can be seen in the table of
contents. Most chapters include subsections titled “Chapter
outline”, “When is [this technique] used”,
“Data example”, “Basic concepts”, “Discussion
of computer programs”, “What to watch out for”,
“Summary”, and “Problems”.
The UCLA website,
http://www.ats.ucla.edu/stat/examples/cama4
is another resource for readers of this book. Here many of the
examples that were in the fourth edition of the book
are demonstrated in Stata and in four other statistical packages.
The data for the fifth edition are available for download from within Stata
so that you can practice applying the techniques as you read.
If you are looking for derivations and proofs, this book is not for you. If
you are looking for guidance on techniques to use, when to use them, and how
to interpret what they produce, this book will prove helpful.
Table of contents
Preface
Author’s Biographies
I. Preparation for Analysis
1. What is multivariate analysis?
1.1 Defining multivariate analysis
1.2 Examples of multivariate analyses
1.3 Multivariate analyses discussed in this book
1.4 Organization and content of the book
2. Characterizing data for analysis
2.1 Variables: their definition, classification, and use
2.2 Defining statistical variables
2.3 Stevens’s classification of variables
2.4 How variables are used in data analysis
2.5 Examples of classifying variables
2.6 Other characteristics of data
2.7 Summary
2.8 Problems
3. Preparing for data analysis
3.1 Processing data so they can be analyzed
3.2 Choice of a statistical package
3.3 Techniques for data entry
3.4 Organizing the data
3.5 Example: depression study
3.6 Summary
3.7 Problems
4. Data screening and transformations
4.1 Transformations, assessing normality and independence
4.2 Common transformations
4.3 Selecting appropriate transformations
4.4 Assessing independence
4.5 Summary
4.6 Problems
5. Selecting appropriate analyses
5.1 Which analyses to perform?
5.2 Why selection is often difficult
5.3 Appropriate statistical measures
5.4 Selecting appropriate multivariate analyses
5.5 Summary
5.6 Problems
II. Applied Regression Analysis
6. Simple regression and correlation
6.1 Chapter outline
6.2 When are regression and correlation used?
6.3 Data example
6.4 Regression methods: fixedX case
6.5 Regression and correlation: variableX case
6.6 Interpretation: fixedX case
6.7 Interpretation: variableX case
6.8 Other available computer output
6.9 Robustness and transformations for regression
6.10 Other types of regression
6.11 Special applications of regression
6.12 Discussion of computer programs
6.13 What to watch out for
6.14 Summary
6.15 Problems
7. Multiple regression and correlation
7.1 Chapter outline
7.2 When are regression and correlation used?
7.3 Data example
7.4 Regression methods: fixedX case
7.5 Regression and correlation: variableX case
7.6 Interpretation: fixedX case
7.7 Interpretation: variableX case
7.8 Regression diagnostics and transformations
7.9 Other options in computer programs
7.10 Discussion of computer programs
7.11 What to watch out for
7.12 Summary
7.13 Problems
8. Variable selection in regression
8.1 Chapter outline
8.2 When are variable selection methods used?
8.3 Data example
8.4 Criteria for variable selection
8.5 A general F test
8.6 Stepwise regression
8.7 Subset regression
8.8 Discussion of computer programs
8.9 Discussion of strategies
8.10 What to watch out for
8.11 Summary
8.12 Problems
9. Special regression topics
9.1 Chapter outline
9.2 Missing values in regression analysis
9.3 Dummy variables
9.4 Constraints on parameters
9.5 Regression analysis with multicollinearity
9.6 Ridge regression
9.7 Summary
9.8 Problems
III. Multivariate Analysis
10. Canonical correlation analysis
10.1 Chapter outline
10.2 When is canonical correlation analysis used?
10.3 Data example
10.4 Basic concepts of canonical correlation
10.5 Other topics in canonical correlation
10.6 Discussion of computer programs
10.7 What to watch out for
10.8 Summary
10.9 Problems
11. Discriminant analysis
11.1 Chapter outing
11.2 When is discriminant analysis used?
11.3 Data example
11.4 Basic concepts of classification
11.5 Theoretical background
11.6 Interpretation
11.7 Adjusting the dividing point
11.8 How good is the discrimination?
11.9 Testing variable contributions
11.10 Variable selection
11.11 Discussion of computer programs
11.12 What to watch out for
11.13 Summary
11.14 Problems
12. Logistic regression
12.1 Chapter outline
12.2 When is logistic regression used?
12.3 Data example
12.4 Basic concepts of logistic regression
12.5 Interpretation: categorical variables
12.6 Interpretation: continuous variables
12.7 Interpretation: interactions
12.8 Refining and evaluating logistic regression
12.9 Nominal and ordinal logistic regression
12.10 Applications of logistic regression
12.11 Poisson regression
12.12 Discussion of computer programs
12.13 What to watch out for
12.14 Summary
12.15 Problems
13. Regression analysis with survival data
13.1 Chapter outline
13.2 When is survival analysis used?
13.3 Data examples
13.4 Survival functions
13.5 Common survival distributions
13.6 Comparing survival among groups
13.7 The loglinear regression model
13.8 The Cox regression model
13.9 Comparing regression models
13.10 Discussion of computer programs
13.11 What to watch out for
13.12 Summary
13.13 Problems
14. Principal components analysis
14.1 Chapter outline
14.2 When is principal components analysis used?
14.3 Data example
14.4 Basic concepts
14.5 Interpretation
14.6 Other uses
14.7 Discussion of computer programs
14.8 What to watch out for
14.9 Summary
14.10 Problems
15. Factor analysis
15.1 Chapter outline
15.2 When is factor analysis used?
15.3 Data example
15.4 Basic concepts
15.5 Initial extraction: principal components
15.6 Initial extraction: iterated components
15.7 Factor rotations
15.8 Assigning factor scores
15.9 Application of factor analysis
15.10 Discussion of computer programs
15.11 What to watch out for
15.12 Summary
15.13 Problems
16. Cluster analysis
16.1 Chapter outline
16.2 When is cluster analysis used?
16.3 Data example
16.4 Basic concepts: initial analysis
16.5 Analytical clustering techniques
16.6 Cluster analysis for financial data set
16.7 Discussion of computer programs
16.8 What to watch out for
16.9 Summary
16.10 Problems
17. Loglinear analysis
17.1 Chapter outline
17.2 When is loglinear analysis used?
17.3 Data example
17.4 Notation and sample considerations
17.5 Tests and models for twoway tables
17.6 Example of a twoway table
17.7 Models for multiway tables
17.8 Exploratory model building
17.9 Assessing specific models
17.10 Sample size issues
17.11 The logit model
17.12 Discussion of computer programs
17.13 What to watch out for
17.14 Summary
17.15 Problems
18. Correlated outcomes regression
18.1 Chapter outline
18.2 When is correlated outcomes regression used?
18.3 Data examples
18.4 Basic concepts
18.5 Regression of clustered data
18.6 Regression of longitudinal data
18.7 Other analyses of correlated outcomes
18.8 Discussion of computer programs
18.9 What to watch out for
18.10 Summary
18.11 Problems
19. Appendix A
A.1 Data sets and how to obtain them
A.2 Chemical companies financial data
A.3 Depression study data
A.4 Financial performance cluster analysis data
A.5 Lung cancer survival data
A.6 Lung function data
A.7 Paternal HIV data
A.8 Northridge earthquake data
A.9 School data
A.10 Mice data
References
Index