Practical Statistics for Data Scientists, 2e: 50+ Essential Concepts Using R and Python Paperback – 29 Jun. 2020

I got this because I am taking a data analytics course that is not explained that well and I need to fill up my gaps in statistics. It is a good book

Content is extremely well written, you'll learn the fundamentals of data science and gain an understanding of how data can be used to model different situations, as well as the mathematical/practical methods to do so.

This book explores traditional statistical concepts (median, correlation, distributions etc) before moving on to machine learning models (the traditional, statistical kind like logistic regression and trees, not neural networks). Both classification and regression tasks are explored.The book is broken down into fairly digestible sections, where each section states the idea, before exploring it with both R and Python snippets and some recurring data sets. Data output is in R.In general, the quality of writing is good and I particularly liked how the authors pointed out where something useful in classical statistics might not be particularly relevant for data science or machine learning (e.g. p-values). The book is very practical in that sense, and I appreciate the more nuanced details about some real-life problems you might face (like the "rare-class" problem).But I think some things could be improved. First, the authors seem keen to state a formal, mathematical explanation of a concept but don't always bring it to life with an example. For someone not trained in stats, that can be a little daunting (they don't state what background knowledge they expect).Second, I think they try to squeeze a bit too much in that isn't really needed. For example, they talk about the F-statistic briefly but almost as a reference. I was left none the wiser as to how I can use it in my work.My suggestions to the authors would therefore be to: bring the concepts to life a bit more and connect more of the dots. Otherwise, a worthwhile book if you are into data science.

The book provides good explanations of complicated issues

I bought this book as someone who had very briefly studied statistics a very long time ago and needed to refresh my knowledge. It is very clearly written and contains a great introduction to the basic concepts presented. It also contains snippets of code in both Python and R to demonstrate how one would go about producing the graphs shown in the book.

It's a good skim through read rather than an in depth book. I I appreciate it being a good beginner friendly book.

Just finished reading it. I don't get the point of this book, especially when all the material (yes, all of it) is already present in other comprehensive books such as Hands on Machine Learning with Tensorflow and Keras and also at a lower price and with much more content related to both statistics and machine learning. Half of the book insists on theoretical aspects related to decision trees, clustering, regression, etc, things already explained much better in other books (see above).O'Reilly, what is going on that lately you keep publishing these books that offer the same contents but with different fancy names? Not to mention that half of the content of most books related to data science and machine learning that you publish use a third of the book's content to write about the same boring introductory things that we we keep seeing in each of the fancy titled book you release, and Practical Statistics for Data Scientists is no exception to the rule. Machine Learning Design Patterns and Data Science on AWS are two examples of very succinct and on-point books which perhaps you could use as a baseline for publishing future data science books.For future buyers, if you think about getting this book, I suggest getting Hands on Machine Learning with Tensorflow and Keras (2nd or 3rd edition) because it already contains every single aspect contained in Practical Statistics for Data Scientists and many more chapters related to statistics and data science.