266 Pages
    by Routledge

    266 Pages
    by Chapman & Hall

    Repeated measures data arise when the same characteristic is measured on each case or subject at several times or under several conditions. There is a multitude of techniques available for analysing such data and in the past this has led to some confusion. This book describes the whole spectrum of approaches, beginning with very simple and crude methods, working through intermediate techniques commonly used by consultant statisticians, and concluding with more recent and advanced methods. Those covered include multiple testing, response feature analysis, univariate analysis of variance approaches, multivariate analysis of variance approaches, regression models, two-stage line models, approaches to categorical data and techniques for analysing crossover designs. The theory is illustrated with examples, using real data brought to the authors during their work as statistical consultants.

    Acknowledgements -- Preface -- 1 Introduction -- 1.1 Background and objectives -- 1.2 Overview -- 2 Some simple analyses -- 2.1 Comparisons at individual times -- 2.2 Response feature analysis -- 2.3 Individual curve fitting -- 2.4 Further reading -- 3 Univariate analysis of variance -- 3.1 The fundamental model -- 3.2 Anova -- 3.3 Calculation of expected mean-squares -- 3.4 Expected mean-squares by 'synthesis' -- 3.5 Contrasts, compound symmetry and F-tests -- 3.6 Relaxing assumptions: univariate, modified univariate, or multivariate tests? -- 3.7 Further reading -- 4 Multivariate analysis -- 4.1 Models without special covariance structure -- 4.2 Hotelling's T2 -- 4.3 Testing for polynomial trends -- 4.4 Manova -- 4.5 Further reading -- 5 Regression models -- 5.1 Special case -- 5.2 General case -- 5.3 Structured covariance case -- 5.4 Some covariance structures -- 5.5 Further reading -- 6 Two-stage linear models -- 6.1 Random regression coefficients -- 6.2 Estimation and testing -- 6.3 Particular aspects -- 6.4 Examples -- 6.5 Further reading -- 7 Crossover experiments -- 7.1 Simple 2 x 2 designs -- 7.2 A Bayesian approach to 2 x 2 designs -- 7.3 More complex crossover designs for two treatments -- 7.4 Crossover trials with a binary response -- 7.5 Further topics -- 7.6 Further reading -- 8 Categorical data -- 8.1 Introduction -- 8.2 Markov chain models -- 8.3 Log-linear models -- 8.4 Linear model methods for group and time comparisons -- 8.5 Randomization test approaches -- 8.6 Some special cases -- 8.7 Further reading -- 9 Some further topics -- 9.1 Some practical matters -- 9.2 Antedependence -- 9.3 Tracking -- 9.4 Nonlinear growth curves -- 9.5 Non-normal observations -- 10 Computer software and examples -- 10.1 Repeated measures facilities in BMDP, SPSS' and SAS -- 10.2 Example 1 - BMDP program 2V -- 10.3 Example 2 - BMDP program 2V -- 10.4 Example 3 - BMDP program 2V -- 10.5 Example 4 - SPSS' program MANOVA -- 10.6 Example 5 - SPSS' program MANOVA -- 10.7 Example 6- BMDP program 5V -- Bibliography /Supplementary References -- Index.

    Biography

    Crowder, Martin J. | Hand, David J.

    "...useful in a consultant statistician's work as well as for a student's statistical computer exercises and can be recommended for both cases."
    -Statistics