Time Series Analysis with Python Cookbook: Practical recipes for exploratory data analysis, data preparation, forecasting, and model evaluation

This book provides a comprehensive Python code bank and reference manual for performing time series analysis and forecasting confidently. It covers techniques such as statistical, machine learning, and deep learning algorithms, as well as evaluation, diagnosis, and optimization methods. The book covers practical techniques for working with complex data, including data preparation, handling missing data, time zones, and custom business days, anomaly detection, forecasting, and model evaluation. It is suitable for data analysts, business analysts, data scientists, data engineers, and Python developers who want to learn practical Python recipes for time series analysis and forecasting.

Format: Paperback / softback
Length: 630 pages
Publication date: 30 June 2022
Publisher: Packt Publishing Limited

This comprehensive Python code bank and reference manual provide you with the essential tools and knowledge to confidently perform time series analysis and forecasting. Whether you're a novice or an experienced data scientist, this book will empower you to explore forecasting and anomaly detection techniques using statistical, machine learning, and deep learning algorithms.

With its extensive coverage, you'll learn different techniques for evaluating, diagnosing, and optimizing your models, ensuring that you make informed decisions based on your time series data.

The book begins by introducing you to the world of time series data, highlighting its unique characteristics and challenges. You'll learn how to ingest time series data from various sources and formats, including private cloud storage, relational databases, non-relational databases, and specialized time series databases such as InfluxDB.

Next, you'll delve into strategies for handling missing data, dealing with time zones and custom business days, and detecting anomalies using intuitive statistical methods. As you progress, you'll explore more advanced unsupervised ML models, such as Autoencoders and Generative Adversarial Networks (GANs), to uncover hidden patterns and relationships in your data.

The book also covers forecasting using classical statistical models such as Holt-Winters, SARIMA, and VAR. You'll learn practical techniques for handling non-stationary data, using power transforms, ACF and PACF plots, and decomposing time series data with multiple seasonal patterns.

Furthermore, you'll explore ML and DL models using TensorFlow and PyTorch, allowing you to harness the power of these advanced frameworks for time series analysis and forecasting.

Throughout the book, you'll find numerous recipes and code snippets that demonstrate the practical application of the techniques discussed. These recipes will walk you through step-by-step processes, providing you with the tools and knowledge to build robust and effective time series models.

By the end of this book, you'll have a solid understanding of what makes time series data different from other data types and how to apply various imputation techniques to handle missing data. You'll also be equipped with the skills to evaluate, compare, and optimize models, ensuring that you select the most suitable approach for your specific time series analysis and forecasting needs.

Whether you're working in finance, industry, or any other field that relies on time series data, this Python code bank and reference manual will be your go-to resource for achieving accurate and reliable time series analysis and forecasting. So why wait? Start your journey towards becoming a time series expert today!

Dimension: 93 x 75 (mm)
ISBN-13: 9781801075541