In the world of data and analytics, upgrading to the constantly changing requirement is not a need but a necessity. With ever-changing technology, techniques and implementation strategies, most sources of knowledge fail to keep up with the changes which leaves analysts and engineers stranded in time.
However, there are very few books that are perpetual and are the repository of knowledge for data analysts and data engineers anyday. These books are focused towards people who play with data analytics, right from collating it to, cleansing, preparing and doing transformations.
These books are listed in no particular order, let’s take a look at them
How to Measure Anything, a 433 page classic written by Douglas W Hubbard emphasizes strictly on measuring the data only if you will be doing something useful with it. Well, that summarizes the book. However, the sole purpose of measuring things, which may or may not showcase the exact value but to attain a level of understanding on the value in order to make an informed decision is what this book is all about.
We humans are not as good as we think when it comes to predictions, estimating something to its near accurate value. The way you perceive things would take a different route and the ability to use well thought through assumptions and estimates improves significantly as per most readers.
As this book is all about measurement, it answers most of your questions say, what could be the retention rate if there’s a change in the brand of any specific product? This book consists of such techniques which allows to approach the business in a different perspective than normal.
Statistics Done Wrong, a small read with just 176 pages yet power packed with the statistical fallacies, written by Alex Reinhart. From a book this crisp, it might not serve you the best if you are looking to start from the basics of statistics. The author jumps straight into the practices that could go on to be a disaster.
This book is not actually a rant as the title suggests, it's more of the unintentional errors committed by people who work on statistics and this being explained in a subtle manner. The book consists of piles of examples with explanations on how wrong statistics can lead to dithering one’s business instincts.
Despite getting all the theory right, practical implementation of statistics is very hard. Statistics might seem easy on the eye, that’s how loads of research ends up negatively. Statistics done wrong is a guide for what not to do.
Data Science and Big Data Analytics is not a “just another data science book”, this 432 page delight which educates the readers about what Big Data is and how to make the best use of it. The book is a bit vague yet gives great information on all the high-level concepts like randomization, sampling, distribution, sample bias, etc.
Right from understanding big data and drawing analytics from it to the techniques to analyze structured and unstructured data, machine learning implementations, data visualizations is what this book speaks about.
Well articulated concepts with theoretical and practical support (examples are provided in R) and supplemented with visual interpretations. However, this data science book is for data science beginners, much like a big data analytics book for beginners. As per most readers, it’s a preferable handbook to glance at before exams or interviews.
Practical Statistics for Data Scientists is more like a consolidated record of how the basics of statistics are seen from the data scientist’s perspective written by Peter and Andrew Bruce which has a total of 320 pages.
To make the model easily understandable to programmers, python is used but for non-programmers, a python course first would make things easy. The book defines the differences between statistics and data science first, the traditional statistics and how different it is from data science. Not just that, the relation between the two terminologies are also explained.
The book suggests “50 essential concepts” in its title but as per the readers, it’s a lot more than that, the practical examples for resampling, A/B testing, etc and R code language makes this book rich. A must read data science book for beginners.
An Introduction to Statistical Learning has about 426 pages written by Gareth M. James, Daniela Witten, Trevor Hastie, Robert Tibshirani on the vastness of data sets and how statistics is an essential tool to derive the best sense from it.
The topics covered are modeling and prediction techniques along with its applications. In addition to this, linear regression, classification, resampling methods, shrinkage approaches, etc are also explained in-depth.
As per readers, the best thing about the books is that it facilitates to practice at the end of every chapter as every chapter consists of one specific algorithm with R code. This alters perspectives and allows readers to practise these techniques in real life data. A book highly recommended for beginners with basic statistics knowledge and are looking to kick start their career in data science and data analytics.