Statistical Foundations of Data Science offers a comprehensive overview of popular statistical models, modern statistical machine learning methods, and their underlying mathematical principles and theories. This textbook is designed for graduate students and as a research reference on topics such as high-dimensional statistics, sparsity, covariance learning, machine learning, and statistical inference. It contains numerous exercises that combine both theoretical concepts and practical applications.
The ebook kicks off with an exploration of the key characteristics of big data and how these features influence statistical analysis. Following this, it delves into multiple linear regression and elaborates on model-building techniques through nonparametric regression and kernel methods. It extensively covers approaches to sparsity and model selection in contexts such as multiple regression, generalized linear models, quantile regression, robust regression, and hazards regression. Additionally, it thoroughly discusses high-dimensional inference and feature screening.
The textbook also includes an in-depth examination of high-dimensional covariance estimation, the learning of latent factors and hidden structures, along with their uses in statistical estimation, inference, prediction, and machine learning tasks. Moreover, it thoroughly presents statistical machine learning theories and techniques applicable to classification, clustering, and prediction tasks. These methods encompass CART, random forests, boosting, support vector machines, various clustering algorithms, sparse PCA, and deep learning.
978-1466510845, 978-1466510852, 978-0429527616, 978-0367512620, 978-0429542312, 978-0429096280, 978-1032941752
NOTE: This only consists of the eBook Statistical Foundations of Data Science, 1st Edition, in the original PDF format. No access codes are included.
