Rank-Based Methods for Shrinkage and Selection: Wi th Application to Machine Learning

Robust statistics is an important field in contemporary mathematics and applied statistical methods, and Rank-Based Methods for Shrinkage and Selection: With Application to Machine Learning provides a practical guide to producing higher quality data analysis in shrinkage and subset selection. It covers rank theory, penalized rank estimators, ridge, LASSO, Enet, logistic regression, and neural networks, with problem sets to demonstrate their use in machine learning.

Format: Hardback
Length: 480 pages
Publication date: 11 March 2022
Publisher: John Wiley and Sons Ltd

Robust statistics is a vital field in contemporary mathematics and applied statistical methods, playing a crucial role in data analysis and interpretation. In this comprehensive guide, Rank-Based Methods for Shrinkage and Selection: With Application to Machine Learning delves into the theory and methodology of statistical estimation based on rank, offering practical insights for producing higher quality data analysis.

The book begins by introducing the concept of rank and its significance in statistical estimation. It then explores various rank-based methods for shrinkage and selection, including techniques such as ridge regression, LASSO (least absolute shrinkage and selection operator), and Enet (elastic net). These methods aim to obtain parsimonious models with outlier-free prediction, which are valuable in various fields such as economics, biology, and machine learning.

One of the key strengths of this book is its practical and hands-on approach. It provides detailed explanations of the theory and methodology, accompanied by numerous examples and R code demonstrations. This makes it accessible to statisticians, economists, biostatisticians, data scientists, and graduate students who are interested in learning and applying rank-based methods in their research.

The book further elaborates on rank-based theory and application in machine learning. It discusses the development of rank theory and its application to robustify the least squares methodology, which is commonly used in linear regression. The book also introduces penalized rank estimators, such as ridge regression and LASSO, which are designed to handle data with missing values or outliers. These estimators are particularly useful in high-dimensional data analysis and machine learning applications.

In addition to these theoretical discussions, the book covers various topics related to robust data science, including Liu regression, high-dimension, and AR(p) models. It also introduces novel rank-based logistic regression and neural networks, which have shown promising results in various machine learning tasks.

To further enhance the learning experience, the book includes problem sets that allow readers to apply the techniques learned in the book to real-world data sets. These problem sets provide hands-on practice and help readers to understand the practical implications of rank-based methods in data analysis.

Overall, Rank-Based Methods for Shrinkage and Selection: With Application to Machine Learning is a valuable resource for anyone interested in robust statistical estimation and machine learning. It provides a comprehensive and up-to-date introduction to the theory and methodology of rank-based methods, accompanied by practical examples and problem sets. This book will be of great interest to statisticians, economists, biostatisticians, data scientists, and graduate students seeking to enhance their data analysis skills and apply them in their research.

Weight: 804g
Dimension: 239 x 152 x 34 (mm)
ISBN-13: 9781119625391