Applied Multivariate Statistical Analysis

Now in its sixth edition, this textbook presents the tools and concepts used in multivariate data analysis in a style accessible for non-mathematicians and practitioners. Each chapter features hands-on exercises that showcase applications across various fields of multivariate data analysis. These exercises utilize high-dimensional to ultra-high-dimensional data, reflecting real-world challenges in big data analysis. For this new edition, the book has been updated and revised and now includes new chapters on modern machine learning techniques for dimension reduction and data visualization, namely locally linear embedding, t-distributed stochastic neighborhood embedding, and uniform manifold approximation and projection, which overcome the shortcomings of traditional visualization and dimension reduction techniques. Solutions to the book’s exercises are supplemented by R and MATLAB or SAS computer code and are available online on the Quantlet and Quantinar platforms. Practical exercises from this book and their solutions can also be found in the accompanying Springer book by W.K. Härdle and Z. Hlávka: Multivariate Statistics - Exercises and Solutions.

Table of Contents:
Part I Descriptive Techniques.- 1 Comparison of Batches.- Part II Multivariate Random Variables.- 2 A Short Excursion into Matrix Algebra.- 3 Moving to Higher Dimensions.- 4 Multivariate Distributions.- 5 Theory of the Multinormal.- 6 Theory of Estimation.- 7 Hypothesis Testing.- Part III Multivariate Techniques.- 8 Regression Models.- 9 Variable Selection.-10 Decomposition of Data Matrices by Factors.- 11 Principal Components Analysis.- 12 Factor Analysis.- 13 Cluster Analysis.- 14 Discriminant Analysis.- 15 Correspondence Analysis.- 16 Canonical Correlation Analysis.- 17 Multidimensional Scaling.- 18 Conjoint Measurement Analysis.- 19 Applications in Finance.- 20 Computationally Intensive Techniques.- 21 Locally Linear Embedding.- 22 Stochastic Neighborhood Embedding.- 23 Uniform Manifold Approximation and Projection.- Part IV Appendix.- A Symbols and Notations.- B Data.- Index.

About the Author :
Wolfgang Karl Härdle is the Ladislaus von Bortkiewicz Professor of Statistics at the Humboldt-Universität zu Berlin, Germany. He is also a Professor at the Faculty of Mathematics and Physics at the Charles University in Prague, Czech Republic. He teaches quantitative finance and semi-parametric statistics. His research focuses on modern machine learning, multivariate statistics in finance and computational statistics. He is an elected member of the ISI (International Statistical Institute) and the director of the Institute of Digital Assets, Academy of Economic Sciences, Bucharest, Romania. Léopold Simar is an Emeritus Professor of Statistics at UCLouvain, Louvain-la-Neuve, Belgium. He has taught mathematical statistics, multivariate analysis, bootstrap methods in statistics and econometrics at several European universities. His research focuses on non-parametric and semi-parametric methods and bootstrap techniques in statistics and econometrics. He is an elected member of the ISI and the past president of the Belgian Statistical Society and is a regular Visiting Professor at the Sapienza University of Rome, Italy and at the Toulouse School of Economics, France. Matthias R. Fengler is a Professor of Econometrics at the School of Economics and Political Science at the University of St. Gallen, Switzerland. His area of specialization is Financial Econometrics and he works in asset pricing, volatility modeling, risk-management, and the analysis of financial text data. In collaboration with colleagues from both academia and industry, he initiated the University of St. Gallen's Impact Award-winning project Monitoring Consumption Switzerland, which analyzes and explores private debit and credit card expenditures and payment behavior in Switzerland.