Numerous statistical applications are time dependent. Virtually all biomedical, pharmaceutical, and industrial experiments demand repeated measurements over time. The same holds true for market research and analysis. Yet conventional methods, such as the Repeated Measures Analysis of Variance (Rm ANOVA), do not always yield exact solutions, obliging practitioners to settle for asymptotic results and approximate solutions. Generalized inference in Multivariate Analysis of Variance (MANOVA), mixed models, and growth curves offer exact methods of data analysis under milder conditions without deviating from the conventional philosophy of statistical inference.
Generalized Inference in Repeated Measures is a concise, self-contained guide to the use of these innovative solutions, presenting them as extensions of–rather than alternatives to–classical methods of statistical evaluation. Requiring minimal prior knowledge of statistical concepts in the evaluation of linear models, the book provides exact parametric methods for each application considered, with solutions presented in terms of generalized p-values. Coverage includes:
- New concepts in statistical inference, with special focus on generalized p-values and generalized confidence intervals
- One-way and two-way ANOVA, in cases of equal and unequal variances
- Basic and higher-way mixed models, including testing and estimation of fixed effects and variance components
- Multivariate populations, including basic inference, comparison, and analysis of variance
- Basic, widely used repeated measures models including crossover designs and growth curves
With a comprehensive set of formulas, illustrative examples, and exercises in each chapter, Generalized Inference in Repeated Measures is ideal as both a comprehensive reference for research professionals and a text for students.