Nonlinear Multivariate Analysis (Wiley Series in Probability and Statistics)

Nonlinear Multivariate Analysis by Albert Gifi, published by Wiley in July 1990, is a comprehensive resource that explores multivariate analysis techniques applicable to statistical data across various measurement levels, including nominal, ordinal, and interval. This edition spans 602 pages and provides an in-depth examination of methods for assessing the stability of these techniques, incorporating resampling methods such as bootstrap and jackknife, as well as sensitivity analysis through first-order approximations.

Readers will find a detailed discussion of the alternating least squares algorithm as a general model, along with significant special cases. The book covers a range of techniques, including correspondence analysis, principal components analysis, and canonical correlation analysis, while also addressing various ordination methods. Historical overviews, theoretical results, and algorithmic efficiency are included, with a focus on graphical displays and the nonlinear transformation of variables to enhance data visualization.