A new and excellent book on Statistical Models in Python
The new book "Building Statistical Models in Python" by Huy Hoang Nguyen, Paul N Adams, and Stuart J Miller is an accessible and comprehensive book on statistical models in Python that covers a lot of ground and gives the much needed strong foundations in statistics that many professional nowadays need, including data scientists and data professionals.
It starts with an insightful introduction to statistics covering population versus sample, inference from samples, and sampling strategies. It then goes on to cover data distributions of the different types, also covering normal distribution, the central limit theorem, and bootstraping.
A whole chapter, the third, is dedicated to hypothesis testing followed by parameteric tests in the following chapter. Regression (simple and multiple linear) and classification models are also covered.
I especially liked the covering of feature selection (not covered in many books properly) and shrinkage methods. Dimension reduction is also explained clearly with an a hands-on example. Different from many other titles, linear discriminant analysis also receives a very good treatment.
Being comprehensive, this book also covers statistical time series models starting from the first concepts and then having a whole chapter dedicated to ARIMA models and another to multivariate time series. Afterwards, survival analysis is well covered in two chapters.
The authors show a solid grasp of the subject, including the underlying mathematics, which gives them the opportunity to provide unique insights and a clear exposition. Unfortunately many books on statistical models with Python, including those fully dedicated to time series, don’t have the same excellent and clear coverage, lacking a robust understanding of the underlying mathematics.
What is impressive in this book is how much their authors cover and with such a clarity and dedication. I highly recommend it.
Please use the link below to order it on Amazon.com
