The Unobserved: Latent Variable Models for Multivariate Data

Preface

This book is under construction.

We are pleased to present this book on the use of latent variable models and probabilistic models for multivariate data. As authors, we bring a wealth of experience in this field, having both completed our PhDs in statistics with a focus on the application of these techniques.

During our doctoral studies, we gained a deep understanding of the power and versatility of latent variable models and probabilistic models, and we have since used them extensively in our research. Guillaume’s work focused on the estimation of these complex models in nonlinear settings, while Julien’s work focused on nonparametric estimation using splines. Our combined expertise allows us to offer a comprehensive view of these models from both theoretical and practical perspectives.

In this book, we provide an introduction to the fundamental concepts of latent variable models and probabilistic models, along with practical guidance on how to implement them. We cover a wide range of topics, including factor analysis, variational autoencoders, and generative adversarial networks, and illustrate their application in various fields, from biology to economics.

Our aim in writing this book is to share our knowledge and experience with others who are interested in multivariate data analysis, and to provide a practical guide for those who want to use these powerful techniques in their own research. We hope that this book will be a valuable resource for students, researchers, and practitioners alike, and we welcome any feedback or suggestions for future editions.