What do you think?
Rate this book
Statistical Regression and Classification: From Linear Models to Machine Learning
Statistical Regression and From Linear Models to Machine Learning takes an innovative look at the traditional statistical regression course, presenting a contemporary treatment in line with today's applications and users. The text takes a modern look at
* A thorough treatment of classical linear and generalized linear models, supplemented with introductory material on machine learning methods.
* Since classification is the focus of many contemporary applications, the book covers this topic in detail, especially the multiclass case.
* In view of the voluminous nature of many modern datasets, there is a chapter on Big Data.
* Has special Mathematical and Computational Complements sections at ends of chapters, and exercises are partitioned into Data, Math and Complements problems.
* Instructors can tailor coverage for specific audiences such as majors in Statistics, Computer Science, or Economics.
* More than 75 examples using real data.
The book treats classical regression methods in an innovative, contemporary manner. Though some statistical learning methods are introduced, the primary methodology used is linear and generalized linear parametric models, covering both the Description and Prediction goals of regression methods. The author is just as interested in Description applications of regression, such as measuring the gender wage gap in Silicon Valley, as in forecasting tomorrow's demand for bike rentals. An entire chapter is devoted to measuring such effects, including discussion of Simpson's Paradox, multiple inference, and causation issues. Similarly, there is an entire chapter of parametric model fit, making use of both residual analysis and assessment via nonparametric analysis.
Norman Matloff is a professor of computer science at the University of California, Davis, and was a founder of the Statistics Department at that institution. His current research focus is on recommender systems, and applications of regression methods to small area estimation and bias reduction in observational studies. He is on the editorial boards of the Journal of Statistical Computation and the R Journal . An award-winning teacher, he is the author of The Art of R Programming and Parallel Computation in Data With Examples in R, C++ and CUDA .
* A thorough treatment of classical linear and generalized linear models, supplemented with introductory material on machine learning methods.
* Since classification is the focus of many contemporary applications, the book covers this topic in detail, especially the multiclass case.
* In view of the voluminous nature of many modern datasets, there is a chapter on Big Data.
* Has special Mathematical and Computational Complements sections at ends of chapters, and exercises are partitioned into Data, Math and Complements problems.
* Instructors can tailor coverage for specific audiences such as majors in Statistics, Computer Science, or Economics.
* More than 75 examples using real data.
The book treats classical regression methods in an innovative, contemporary manner. Though some statistical learning methods are introduced, the primary methodology used is linear and generalized linear parametric models, covering both the Description and Prediction goals of regression methods. The author is just as interested in Description applications of regression, such as measuring the gender wage gap in Silicon Valley, as in forecasting tomorrow's demand for bike rentals. An entire chapter is devoted to measuring such effects, including discussion of Simpson's Paradox, multiple inference, and causation issues. Similarly, there is an entire chapter of parametric model fit, making use of both residual analysis and assessment via nonparametric analysis.
Norman Matloff is a professor of computer science at the University of California, Davis, and was a founder of the Statistics Department at that institution. His current research focus is on recommender systems, and applications of regression methods to small area estimation and bias reduction in observational studies. He is on the editorial boards of the Journal of Statistical Computation and the R Journal . An award-winning teacher, he is the author of The Art of R Programming and Parallel Computation in Data With Examples in R, C++ and CUDA .
528 pages, Kindle Edition
Published September 19, 2017
6 people are currently reading
17 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 of 1 review
March 22, 2018
I bought this book sight-unseen, just because I had a very favorable impression of the author's prior work, Art of R Programming. This is what I found:
* Although the R language is not explicitly mentioned in the book's title or subtitle, the book does make heavy use of R code and examples.
* The book is organized in what I would call a "multi-pass" approach. Instead of building up the concepts in order of dependency, it first gives an overview of the space, with plenty of forward references, and then goes back and fills in the details. You may or may not like this approach, but given the choice to do it this way, I think it's well-executed.
* The book assumes that the reader is a programmer, and thus will find code examples a useful form of explication. You probably want to be at least a journeyman-level R programmer in order for this to be helpful for you.
* The treatment of the classical assumptions in linear regression is quite intuitive and highlights the practical importance (or lack thereof).
* The production seems a bit hurried. The computer code is not well typeset, and there are some typos and other errors. The author has been very responsive and is currently putting together some errata on his web page, so I suggest checking that for updates. Search for "matloff regclass"
The copy I got had some printing issues (color illustrations were in B&W) but the author and publisher checked other copies and they don't have this problem, so it seems I just got a lemon and yours will probably be fine. Check your book for misprints when you get it.
Overall a worthwhile read.
* Although the R language is not explicitly mentioned in the book's title or subtitle, the book does make heavy use of R code and examples.
* The book is organized in what I would call a "multi-pass" approach. Instead of building up the concepts in order of dependency, it first gives an overview of the space, with plenty of forward references, and then goes back and fills in the details. You may or may not like this approach, but given the choice to do it this way, I think it's well-executed.
* The book assumes that the reader is a programmer, and thus will find code examples a useful form of explication. You probably want to be at least a journeyman-level R programmer in order for this to be helpful for you.
* The treatment of the classical assumptions in linear regression is quite intuitive and highlights the practical importance (or lack thereof).
* The production seems a bit hurried. The computer code is not well typeset, and there are some typos and other errors. The author has been very responsive and is currently putting together some errata on his web page, so I suggest checking that for updates. Search for "matloff regclass"
The copy I got had some printing issues (color illustrations were in B&W) but the author and publisher checked other copies and they don't have this problem, so it seems I just got a lemon and yours will probably be fine. Check your book for misprints when you get it.
Overall a worthwhile read.
Displaying 1 of 1 review