Statistical Learning with Sparsity: The Lasso and Generalizations (Chapman & Hall/CRC Monographs on Statistics and Applied Probability) 1st Edition

This book is about bet on sparsity. In Machine Learning, there are plenty of approaches that might work on data of interest. The accent is on cases p>>n and one wants to get more interpretable models.

Very helpful!! Exactly what I need, how I wish I could have found it earlier.

Awesome! This book is right on time. But it. Read it.

Great book!

Great product, fast delivery, thanks!

Though not mentioned as often as that buzz term “Big Data” it’s also the case that, just as the collection of data is expanding exponentially, so is the number of data modeling techniques. In fact, it’s growing so rapidly that it’s hard for even the practicing statistician to keep track of it all.One of the techniques which is becoming standard practice in predictive modeling and causal inference is the lasso. Developed around the turn of the last century, the lasso is designed to handle data generation under sparsity where sparsity can be loosely defined as having many inputs with little or no effect on the target variable.For example, one could be trying to predict users’ preferences on Netflix and have each individual movie as a predictor. Obviously, many of the effects would be zero. Being too large for stepwise regression, the lasso and similar techniques are able to solve these problems on twenty-first century computers within minutes if not seconds.And while this book caters to upper level graduate students and was far more than I needed to gain an understanding of the lasso and related techniques (such as the group lasso when one has highly correlated predictors), I strongly recommend the book to fellow applied scientists. Clearly written with ample illustrations and examples the book sets a standard for how a technical text should be written.Given its uncomplicated prose and the importance of the lasso to modern researchers I gave the book five stars since it is a useful guide to social scientists and data scientists alike. Ideally experienced in graduate school, the text is still helpful for those who, like myself, are merely trying to keep up with the latest trends in their disciplines. Highly recommended.

The statistics/machine learning community has been bombarded with so many variants of LASSO, for so many different types of methodology, without any general, unifying treatment of ths subject. The result is more confusion than insight. This book fills that void, and is sure to be much cited as a reference. It will be quite useful to me.

Hastie and Tibshirani are machine learning superstars and I believe this new resource will play an important role in statistical learning just like their previous texts. The timing is perfect for a deep look at the lasso as big data is placing stringent requirements on how enterprise data assets are being used for strategic advantage.

Great book and It was shipped very quickly.

A great book!

Excellent book! Very pedagogical. Gives very clear insights.