High-Dimensional Statistics: A Non-Asymptotic Viewpoint (Cambridge Series in Statistical and Probabilistic Mathematics, Series Number 48) 1st Edition

Naturally a heavy mathematics text will be a little dry, but nevertheless, I think it is well written and readable. I am still working my way through the book (as in pondering the exercises) but I have found it to be a valuable resource so far. If you want to learn about concentration inequalities this is a great place to start.I would also recommend reading "The Elements of Statistical Learning" by Tibsharini, Hastie, and Friedman as well as "Computer Age Statistical Inference" by Bradley Efron.

This book is very well written and covers lots of state of the art topics. It is a great book for graduate students as well as researchers in statistics, machine learning, and electrical engineering. Covers topics on the concentration of measure, entropy, graphical models, RKHS, and lots of other useful topics that can be used as a basis to do novel research. Totally recommend the book. I use this book as one of the references for statistical machine learning I teach.

For econometricians who want to explore high-dimensional methods, this book is a great reference.

This is the best high dimensional statistics book I have ever encountered. A lot of bounds stuff are (supposedly) tedious but Martin has the magic of explaining them well both technically and intuitively. You rarely see "it's easy to see ..." this kind of phrase in this book; all steps are accompanied with very detailed explanations (even the steps that use very obvious inequality, i.e. Markov). And everything that's "left as exercises" are durable. A lot of books are horrible at explaining empirical process / minimax lower bounds; they either fall into the "too math" or "too hand-wavy" kind of trap. But this book is just right. It's rigorous at the best point that does not frustrate you. I strongly recommend this book for everyone interested in doing high dimensional statistics, machine learning theory, compressive sensing and many, many other fields related to high dimension model / data.

We used this book for Stat200C: High dimensional stats at UCLA. We covered chapter 1, 2, 5, 6, 7 and part of 13. Classmates gave presentation on chapter 4, 6, 7, 8 and 11.This is not a easy book but worth effort reading it. It does not require measure theory. Personally, I like chapter 2, 5 and 7 most. Some proofs in chapter 11 are lack of details (e.g. sparsistency of neighborhood-estimator in section 11).If you are a graduate student in stats/biostats , you will love examples in this excellent textbook

The binding condition of the book I got is bad, which is very disappointing because it doesn't look like a new book.

Thank you