Functional Data Analysis (Second Edition)

Author: (UK) Jim Ramsay
Publisher:
Publication Date: 2006-01-01
Features: Today, scientists collect curve samples and other functional observations. This monograph discusses the ideas and techniques for analyzing such data, primarily including classical linear regression methods, principal component analysis, linear modeling, canonical correlation analysis, and special functional techniques such as curve registration and principal differential analysis. The book consistently uses data from real-world applications to introduce the motivation behind the methods and provide illustrative examples. It particularly demonstrates how functional methods can uncover new characteristics of data by discussing the smoothness of the data-generating process. These data primarily come from applications in growth analysis, meteorology, biomechanics, equine science, economics, and medicine. The book presents innovative statistical techniques while ensuring that the mathematical arguments are accessible to most readers. Much of the content is based on the author's own work, with some material appearing in print for the first time. This book is suitable for students, applied data analysis scholars, and researchers, offering significant value for research in statistics and other broad fields.
The book's author Jim Ramsay is a professor of psychology at McGill University, the president of the Statistical Society of Canada, and an international authority in multivariate analysis and many other fields. He has collaborated with researchers in multiple disciplines, including speech clarity, motor control, meteorology, psychology, and human physiology, publishing numerous papers in journals on statistics and its applications, making technical contributions to the field of functional data analysis. The book's second author, Bernard Silverman, is a professor of statistics at the University of Bristol and the author of the well-known book Density Estimation and Data Analysis in Statistics. He is also a co-author of Nonparametric Regression and Generalized Linear Models. Silverman has been awarded the President's Medal by the Council of Presidents of the Statistical Society and two Guy medals from the Royal Statistical Society for his work in smoothing methods, applied statistics, computational statistics, and theoretical statistics.