Computing Reviews

Basic elements of computational statistics
Härdle W., Okhrin O., Okhrin Y., Springer International Publishing,New York, NY,2017. 305 pp.Type:Book
Date Reviewed: 01/24/19

Researchers frequently use statistics to analyze their results. Statistical analysis is a vital tool to confirm or reject hypotheses. In this context, R, one of the most famous programs used for data analysis and statistics, is a powerful, time-saving tool. The goal of this book is to present different univariate and multivariate data analysis techniques.

This book is focused on readers who want to learn statistics, and it uses R as a way to complement the concepts explained. In general, after a concept is explained, the authors present the corresponding R source code. For example, after explaining the Pearson correlation, the authors show how to calculate it in R. Although the first chapter is an introduction to R, it is generally assumed that readers have some prior knowledge of the language.

After introducing the basics of R, the first chapters explain numerical techniques (such as matrix operations, numerical integration, and so on) and the main discrete distributions (binomial, multinomial, and so on). The following chapters focus on univariate distributions and univariate statistical analysis; the main univariate statistical tests are covered. Multivariate distributions are then introduced along with regression models and the main multivariate statistical tests. The final chapters discuss the generation of random numbers and advanced graphical plots in R.

In summary, this book can help readers learn statistics from scratch. However, it is maybe not the best option for researchers or PhD students looking to identify which tests to use in their analyses (other books such as Experimentation in software engineering [1] may be of more help). Also, while the authors claim in the preface that the book’s goal is to present data analysis in a way that is understandable for non-mathematicians, the language used is quite technical and in general there are no other examples besides the R source code. For this reason, I mainly recommend the book to students, researchers, and practitioners wanting to learn more about statistics.


1)

Oudshoorn, M. Review of Experimentation in software engineering, by Claes Wohlin et al. Computing Reviews (Oct. 9, 2012), CR Rev. No. 140582 (1302-0062).

Reviewer:  Santiago Vidal Review #: CR146397 (1904-0095)

Reproduction in whole or in part without permission is prohibited.   Copyright 2024 ComputingReviews.com™
Terms of Use
| Privacy Policy