Year: 2024
Type: article
Abstract: The second edition of Multivariate Analysis by Kanti V. Mardia, John T. Kent, and Charles C. Taylor is a commendable update to the seminal work first published in 1979. The first edition has been widely used by academic staff, students, and statisticians around the world. It was recommended as a textbook when I was a student. Though the basic principles and topics in multivariate analysis remain the same, many developments in multivariate analysis have taken place over the last 40 years, which warranted a revised edition. This new edition not only maintains the rigorous analytical approach that has long been a hallmark of the text but also introduces significant cutting-edge enhancements in the subject, making it an indispensable resource for both students and practitioners in the field of multivariate statistics. Knowledge of multivariate analysis is essential not only for students and researchers in statistics but also for big data analysts and researchers in other disciplines using these tools. A wide range of material in multivariate analysis is covered in the first edition, which has been updated in this edition, including the classical themes of multivariate normal theory, multivariate regression, inference, discriminant analysis, multidimensional scaling, factor analysis, canonical correlation analysis, cluster analysis, principal component analysis, and multivariate analysis of variance. Some new additions in the second edition include a chapter on non-normal distributions, which covers elliptic distributions and copulas. A new chapter on Graphing Models has been included, covering both Gaussian and Log-linear models. The chapter on discriminant analysis (Chapter 12) has been completely updated and includes cutting-edge supervised learning topics like classification trees, neural networks, and support vector machines, in addition to logistic regression models. The chapter on cluster analysis (Chapter 14), described as unsupervised learning, has also been updated to include the latest developments. A new chapter on high-dimensional data (Chapter 16) has been introduced, covering shrinkage methods, ridge regression, Least Absolute Selection and Shrinkage Operator, Partial Least Square regression, etc. The appendix includes a review of matrix algebra, univariate statistics, and R commands and data. With the latest updates to statistical methodologies, and recent developments and advancements in multivariate analysis, the second edition is more relevant to contemporary data analysts. Additionally, the second edition features expanded and updated examples and exercises, making it a perfect textbook for students and an effective teaching tool for instructors. I strongly recommend this as a textbook for undergraduate and postgraduate students. I also recommend it to practicing statisticians and big data analysts who use these multivariate tools frequently in their research. This book will, of course, be a welcome addition to any university library.
Author: Kuldeep Kumar
Institution: Bond University
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4403204804