Analysis of Multivariate and High-Dimensional Data

'Big data' poses challenges that require both classical multivariate methods and contemporary techniques from machine learning and engineering. This modern text equips you for the new world - integrating the old and the new, fusing theory and practice and bridging the gap to statistical learning. The theoretical framework includes formal statements that set out clearly the guaranteed 'safe operating zone' for the methods and allow you to assess whether data is in the zone, or near enough. Extensive examples showcase the strengths and limitations of different methods with small classical data, data from medicine, biology, marketing and finance, high-dimensional data from bioinformatics, functional data from proteomics, and simulated data. High-dimension low-sample-size data gets special attention. Several data sets are revisited repeatedly to allow comparison of methods. Generous use of colour, algorithms, Matlab code, and problem sets complete the package. Suitable for master's/graduate students in statistics and researchers in data-rich disciplines. '... this book is suitable for readers with various backgrounds and interests and can be read at different levels. ... [It] will also be useful for working statisticians who are interested in analysis of multivariate or high-dimensional data.' Yasunori Fujikoshi, Mathematical Reviews 'A major advantage and important feature of this book is that it illustrates the interconnection of various techniques, such as the connection between discriminant analysis and principal component analysis, cluster analysis and principal component analysis, and factor analysis and cluster analysis. The author has undoubtedly taken on an appreciable task of collecting, compiling, connecting and presenting classical as well as recent developments concisely ... Readers are expected to have an adequate background in statistics to understand the topics in this book. This well-designed book is suitable for advanced multivariate courses where the emphasis is more on applications. Application-orientated researchers will also find this book useful. Problems for each chapter are given at the end of each section. Graphics are used at various places to facilitate the understanding of topics. The book is highly recommended for libraries.' Journal of the Royal Statistical Society 'I must highly commend the author for writing an excellent comprehensive review of multivariate and high dimensional statistics ... The lucid treatment and thoughtful presentation are two additional attractive features ... Without any hesitation and with admiration, I would give the author a 10 out of 10 ... The feat she has accomplished successfully for this difficult area of statistics is something very few could accomplish. The wealth of information is enormous and a motivated student can learn a great deal from this book ... I highly recommend [it] to researchers working in the field of high dimensional data and to motivated graduate students.' Ravindra Khattree, International Statistical ReviewPart I. Classical Methods: 1. Multidimensional data; 2. Principal component analysis; 3. Canonical correlation analysis; 4. Discriminant analysis; Part II. Factors and Groupings: 5. Norms, proximities, features, and dualities; 6. Cluster analysis; 7. Factor analysis; 8. Multidimensional scaling; Part III. Non-Gaussian Analysis: 9. Towards non-Gaussianity; 10. Independent component analysis; 11. Projection pursuit; 12. Kernel and more independent component methods; 13. Feature selection and principal component analysis revisited; Index.