Machine Learning for Time Series Forecasting with Python

by Francesca Lazzeri

Learn how to apply the principles of machine learning to  time series modeling with this indispensable resource   Machine Learning for Time Series Forecasting with Python  is an incisive and straightforward examination of one of the most crucial elements of decision-making in finance, marketing, education, and healthcare: time series modeling.   Despite the centrality of time series forecasting, few business analysts are familiar with the power or utility of applying machine learning to time series modeling. Author Francesca Lazzeri, a distinguished machine learning scientist and economist, corrects that deficiency by providing readers with comprehensive and approachable explanation and treatment of the application of machine learning to time series forecasting.  Written for readers who have little to no experience in time series forecasting or machine learning, the book comprehensively covers all the topics necessary to:  Machine Learning for Time Series Forecasting with Python  is full real-world examples, resources and concrete strategies to help readers explore and transform data and develop usable, practical time series forecasts.  Perfect for entry-level data scientists, business analysts, developers, and researchers, this book is an invaluable and indispensable guide to the fundamental and advanced concepts of machine learning applied to time series modeling.     

224 pages, Paperback

Published December 15, 2020

Reviews

[Carlos Uribe · Rating 5 out of 5]

This book fills two important gaps that were missing in most "practical TSA" books of the main editors for years:

1) It provides enough theoretical background for you to get started with TSF on real-life time series problems, but without overwhelming you with unnecessary statistical details/proofs/theorems of the methods that underpin forecasting. While other books are too focused on math and light on real applications, this one is the opposite, so very good news for practitioners.

2) While R has been historically the "only game in town" when it comes to time series analysis (and forecasting), this book uses Python (hurray!). It provides step-by-step instructions and code samples on how to apply Python's scientific stack for time series (statsmodels, scikit-learn, keras, pandas, numpy, etc.), along with advice on best-practices, and tips and tricks that are very time-series-specific.

Apart from these two gaps, I'd say it also covers another topic that is absent in most other books on TSA: forecasting with deep learning models. It has a whole chapter on DL for forecasting (LSTMs and GRUs, mainly) using Keras, with an intro to RNN for newcomers to the DL world, and also with great tips for time-series data-prep specific to RNNs.

Remember I mentioned that this book is very industry-oriented (instead of academic-oriented like many others using R)? The final proof of that is its last chapter, "Model Deployment for Time Series Forecasting", which is dedicated to the final phase of any successful ML project: Productionalization. In this chapter, first you get to learn Azure ML using the Python SDK, and then, taking a use-case of a model developed in earlier chapters as a guiding example, you learn to operationalize TSF models in Azure ML, plus some caveats of MLOps principles specific to time series. As I said, this is a very hands-on book.

TL;DR:
Best choice for practitioners who are relatively new to TSF, who use Python, and who want to make fast (and good) forecasts with their data, with good-enough understanding of what they're coding.

[Kirill · Rating 4 out of 5]

A bit basic, but great python examples

[Anna Maria · Rating 5 out of 5]

I really enjoyed reading this book. The Python examples are great, and the concepts are explained in a clear way!

[Leonardo]

Lo usó De Simone para su tesis.