Computer-Aided Multivariate Analysis, Fourth Edition

by Afifi; Abdelmonem

• ISBN13: 9781584883081
• ISBN10: 1584883081
• Edition: 4th
• Format: Hardcover
• Copyright: 2003-12-29
• Publisher: Chapman & Hall/

Summary

Computer-Aided Multivariate Analysis, Fourth Edition enables researchers and students with limited mathematical backgrounds to understand the concepts underlying multivariate statistical analysis, perform analysis using statistical packages, and understand the output. New topics include Loess and Poisson regression, nominal and ordinal logistic regression, interpretation of interactions in logistic and survival analysis, and imputation for missing values. This book includes new exercises and references, and updated options in the latest versions of the statistical packages. All data sets and codebooks are available for download.The authors explain the assumptions made in performing each analysis and test, how to determine if your data meets those assumptions, and what to do if they do not. What to Watch out for sections in each chapter warn of common difficulties. By reading this text, you will know what method to use with your data set, how to get the results, and how to interpret them and explain them to others.New in the Fourth Edition:· Expanded explanation of checking for goodness of fit in logistic regression and survival analysis· Kaplan-Meier estimates of survival curves, formal tests for comparing survival between groups, interactions and the use of time-dependent covariates in survival analysis· Expanded discussion of how to handle missing values· Latest features of the S-PLUS package in addition to SAS, SPSS, STATA, and STATISTICA for multivariate analysis· Data sets for the problems are available at the CRC web site: http://www.crcpress.com/e_products/downloads/· Commands and output for examples used in the text for each statistical package are available at the UCLA web site: http://www.ats.ucla.edu/stat/examples/cama4/

Author Biography

Abdelmonem Afifi is a Professor in the Department of Biostatistics at the University of California, Los Angeles Virginia A. Clark is a consultant in Sequim, Washington Susanne May is with the Department of Biostatistics at the University of California, San Diego

Table of Contents

Preface	xiii
One Preparation for Analysis	1	(82)
1 What is multivariate analysis?	3	(10)
1.1 Defining multivariate analysis	3	(1)
1.2 Examples of multivariate analyses	3	(3)
1.3 Multivariate analyses discussed in this book	6	(4)
1.4 Organization and content of the book	10	(1)
1.5 References	11	(2)
2 Characterizing data for analysis	13	(10)
2.1 Variables: their definition, classification, and use	13	(1)
2.2 Defining statistical variables	13	(1)
2.3 Stevens's classification of variables	14	(3)
2.4 How variables are used in data analysis	17	(1)
2.5 Examples of classifying variables	18	(1)
2.6 Other characteristics of data	19	(1)
2.7 Summary	19	(1)
2.8 References	20	(1)
2.9 Problems	20	(3)
3 Preparing for data analysis	23	(208)
3.1 Processing data so they can be analyzed	23	(1)
3.2 Choice of a statistical package	24	(2)
3.3 Techniques for data entry	26	(7)
3.4 Organizing the data	33	(6)
3.5 Example: depression study	39	(4)
3.6 Summary	43	(1)
3.7 References	43	(3)
3.8 Problems	46	(3)
4 Data screening and transformations	49	(22)
4.1 Transformations, assessing normality and independence	49	(1)
4.2 Common transformations	49	(4)
4.3 Selecting appropriate transformations	53	(10)
4.4 Assessing independence	63	(1)
4.5 Summary	64	(2)
4.6 References	66	(1)
4.7 Problems	67	(4)
5 Selecting appropriate analyses	71	(14)
5.1 Which analyses to perform?	71	(1)
5.2 Why selection is often difficult	71	(1)
5.3 Appropriate statistical measures	72	(4)
5.4 Selecting appropriate multivariate analyses	76	(3)
5.5 Summary	79	(1)
5.6 References	79	(1)
5.7 Problems	80	(3)
Two Applied Regression Analysis	83	(148)
6 Simple regression and correlation	85	(40)
6.1 Chapter outline	85	(1)
6.2 When are regression and correlation used?	86	(1)
6.3 Data example	86	(2)
6.4 Regression methods: fixed-X case	88	(5)
6.5 Regression and correlation: variable-X case	93	(1)
6.6 Interpretation: fixed-X case	94	(1)
6.7 Interpretation: variable-X case	95	(4)
6.8 Other available computer output	99	(8)
6.9 Robustness and transformations for regression	107	(2)
6.10 Other types of regression	109	(4)
6.11 Special applications of regression	113	(3)
6.12 Discussion of computer programs	116	(1)
6.13 What to watch out for	117	(1)
6.14 Summary	118	(1)
6.15 References	119	(3)
6.16 Problems	122	(3)
7 Multiple regression and correlation	125	(40)
7.1 Chapter outline	125	(1)
7.2 When are regression and correlation used?	126	(1)
7.3 Data example	126	(3)
7.4 Regression methods: fixed-X case	129	(2)
7.5 Regression and correlation: variable-X case	131	(6)
7.6 Interpretation: fixed-X case	137	(3)
7.7 Interpretation: variable-X case	140	(3)
7.8 Regression diagnostics and transformations	143	(5)
7.9 Other options in computer programs	148	(5)
7.10 Discussion of computer programs	153	(5)
7.11 What to watch out for	158	(1)
7.12 Summary	159	(1)
7.13 References	159	(1)
7.14 Problems	160	(5)
8 Variable selection in regression	165	(32)
8.1 Chapter outline	165	(1)
8.2 When are variable selection methods used?	165	(1)
8.3 Data example	166	(4)
8.4 Criteria for variable selection	170	(2)
8.5 A general F test	172	(2)
8.6 Stepwise regression	174	(6)
8.7 Subset regression	180	(3)
8.8 Discussion of computer programs	183	(3)
8.9 Discussion of strategies	186	(3)
8.10 What to watch out for	189	(2)
8.11 Summary	191	(1)
8.12 References	192	(1)
8.13 Problems	193	(4)
9 Special regression topics	197	(34)
9.1 Chapter outline	197	(1)
9.2 Missing values in regression analysis	197	(8)
9.3 Dummy variables	205	(9)
9.4 Constraints on parameters	214	(3)
9.5 Regression analysis with multicollinearity	217	(1)
9.6 Ridge regression	218	(4)
9.7 Summary	222	(1)
9.8 References	223	(2)
9.9 Problems	225	(6)
Three Multivariate Analysis	231	(246)
10 Canonical correlation analysis	233	(16)
10.1 Chapter outline	233	(1)
10.2 When is canonical correlation analysis used?	233	(1)
10.3 Data example	234	(1)
10.4 Basic concepts of canonical correlation	235	(5)
10.5 Other topics in canonical correlation	240	(3)
10.6 Discussion of computer programs	243	(2)
10.7 What to watch out for	245	(1)
10.8 Summary	246	(1)
10.9 References	246	(1)
10.10 Problems	247	(2)
11 Discriminant analysis	249	(32)
11.1 Chapter outline	249	(1)
11.2 When is discriminant analysis used?	250	(1)
11.3 Data example	251	(1)
11.4 Basic concepts of classification	252	(7)
11.5 Theoretical background	259	(2)
11.6 Interpretation	261	(4)
11.7 Adjusting the dividing point	265	(2)
11.8 How good is the discrimination?	267	(3)
11.9 Testing variable contributions	270	(1)
11.10 Variable selection	271	(1)
11.11 Discussion of computer programs	272	(2)
11.12 What to watch out for	274	(1)
11.13 Summary	275	(1)
11.14 References	276	(1)
11.15 Problems	276	(5)
12 Logistic regression	281	(52)
12.1 Chapter outline	281	(1)
12.2 When is logistic regression used?	282	(1)
12.3 Data example	282	(2)
12.4 Basic concepts of logistic regression	284	(1)
12.5 Interpretation: Categorical variables	285	(2)
12.6 Interpretation: Continuous variables	287	(2)
12.7 Interpretation: Interactions	289	(7)
12.8 Refining and evaluating logistic regression	296	(12)
12.9 Nominal and ordinal logistic regression	308	(7)
12.10 Applications of logistic regression	315	(4)
12.11 Poisson Regression	319	(4)
12.12 Discussion of computer programs	323	(1)
12.13 What to watch out for	324	(3)
12.14 Summary	327	(1)
12.15 References	327	(2)
12.16 Problems	329	(4)
13 Regression analysis with survival data	333	(36)
13.1 Chapter outline	333	(1)
13.2 When is survival analysis used?	334	(1)
13.3 Data examples	334	(3)
13.4 Survival functions	337	(6)
13.5 Common survival distributions	343	(1)
13.6 Comparing survival among groups	344	(2)
13.7 The log-linear regression model	346	(2)
13.8 The Cox regression model	348	(11)
13.9 Comparing regression models	359	(3)
13.10 Discussion of computer programs	362	(2)
13.11 What to watch out for	364	(1)
13.12 Summary	365	(1)
13.13 References	365	(2)
13.14 Problems	367	(2)
14 Principal components analysis	369	(22)
14.1 Chapter outline	369	(1)
14.2 When is principal components analysis used?	369	(1)
14.3 Data example	370	(1)
14.4 Basic concepts	371	(4)
14.5 Interpretation	375	(8)
14.6 Other uses	383	(3)
14.7 Discussion of computer programs	386	(1)
14.8 What to watch out for	386	(2)
14.9 Summary	388	(1)
14.10 References	388	(1)
14.11 Problems	389	(2)
15 Factor analysis	391	(26)
15.1 Chapter outline	391	(1)
15.2 When is factor analysis used?	391	(1)
15.3 Data example	392	(1)
15.4 Basic concepts	393	(2)
15.5 Initial extraction: principal components	395	(3)
15.6 Initial extraction: iterated components	398	(4)
15.7 Factor rotations	402	(4)
15.8 Assigning factor scores	406	(2)
15.9 Application of factor analysis	408	(1)
15.10 Discussion of computer programs	409	(3)
15.11 What to watch out for	412	(1)
15.12 Summary	413	(1)
15.13 References	414	(1)
15.14 Problems	415	(2)
16 Cluster analysis	417	(28)
16.1 Chapter outline	417	(1)
16.2 When is cluster analysis used?	417	(2)
16.3 Data example	419	(1)
16.4 Basic concepts: initial analysis	419	(7)
16.5 Analytical clustering techniques	426	(6)
16.6 Cluster analysis for financial data set	432	(5)
16.7 Discussion of computer programs	437	(3)
16.8 What to watch out for	440	(1)
16.9 Summary	440	(1)
16.10 References	441	(1)
16.11 Problems	442	(3)
17 Log-linear analysis	445	(32)
17.1 Chapter outline	445	(1)
17.2 When is log-linear analysis used?	445	(1)
17.3 Data example	446	(2)
17.4 Notation and sample considerations	448	(2)
17.5 Tests and models for two-way tables	450	(4)
17.6 Example of a two-way table	454	(2)
17.7 Models for multiway tables	456	(3)
17.8 Exploratory model building	459	(6)
17.9 Assessing specific models	465	(1)
17.10 Sample size issues	466	(2)
17.11 The logit model	468	(2)
17.12 Discussion of computer programs	470	(1)
17.13 What to watch out for	471	(2)
17.14 Summary	473	(1)
17.15 References	474	(1)
17.16 Problems	475	(2)
Appendix A	477	(4)
A.1 Data sets and how to obtain them	477	(1)
A.2 Chemical companies financial data	477	(1)
A.3 Depression study data	477	(1)
A.4 Financial performance cluster analysis data	478	(1)
A.5 Lung cancer survival data	478	(1)
A.6 Lung function data	478	(1)
A.7 Parental HIV data	479	(2)
Index	481