book

Time Series Analysis with Python Cookbook - Second Edition

Name: Time Series Analysis with Python Cookbook - Second Edition
Author: Tarek A. Atwan
ISBN: 9781805124283

by Tarek A. Atwan

January 2026

Intermediate to advanced

812 pages

21h 41m

English

Packt Publishing

Read now

Unlock full access

Preface
Who this book is forWhat this book coversTo get the most out of this bookGet in touchFree Benefits with Your BookHow to Unlock
Reading Time Series Data from Files
Free Benefits with Your BookTechnical requirementsReading data from CSV and other delimited filesGetting readyHow to do it…How it works…There’s more…See alsoReading data from an Excel fileGetting readyHow to do it…How it works…There’s more…See alsoReading data from URLsGetting readyHow to do it…Reading data from GitHubReading data from a public S3 bucketReading data from a private S3 bucketReading data from HTMLHow it works…There’s more…See alsoReading data from Parquet filesGetting readyHow to do it…Reading all partitionsReading specific partitionsHow it works…There’s more…See alsoGet This Book’s PDF Version and Exclusive Extras
Reading Time Series Data from Databases
Reading data from a relational databaseGetting readyHow to do it…Using psycopgUsing pandas and SQLAlchemyHow it works…There’s more…Generating the connection URL in SQLAlchemyExtending to the Amazon Redshift databaseChunking with pandasSee alsoReading data from SnowflakeGetting readyHow to do it...Using Snowflake Python ConnectorUsing SQLAlchemyUsing the Snowpark APIHow it works...Snowflake Python ConnectorSQLAlchemy APIThe Snowpark APIThere’s more...See alsoReading data from MongoDBGetting readyHow to do it…How it works…Establishing the MongoDB connectionNavigating databases and collectionsQuerying documentsFiltering data with the MongoDB Query Language (MQL)Working with Time Series dataThere’s more…See also
Persisting Time Series Data to Files
Technical requirementsSerializing time series data with pickleGetting readyHow to do it…Writing to a pickle file using pandasWriting a pickle file using the pickle libraryHow it works…There’s more…See alsoWriting to CSV and other delimited filesGetting readyHow to do it…How it works…There’s more…See alsoWriting data to an Excel fileGetting readyHow to do it…How it works…There’s more…See alsoStoring data in cloud storage (AWS, GCP, and Azure)Getting readyHow to do it…Writing to Amazon S3 using pandasWriting to Google Cloud Storage using pandasWriting to Azure Blob Storage using pandasHow it works…There’s more…Writing to Amazon S3 using the boto3 libraryWriting to Google Cloud Storage using the google-cloud-storage libraryWriting to Azure Blob Storage using the azure-storage-blob librarySee alsoGet This Book’s PDF Version and Exclusive Extras
Persisting Time Series Data to Databases
Technical requirementsWriting time series data to a relational databaseGetting readyHow to do it…How it works…There’s more…Writing to Amazon Redshift with SQLAlchemyWriting to Amazon Redshift using redshift_connectorWriting to Amazon Redshift with AWS SDK for pandasSee alsoWriting time series data to MongoDBGetting readyHow to do it…How it works…There’s more…See alsoWriting time series data to SnowflakeGetting readyHow to do it…Using the Snowflake Python connectorUsing SQLAlchemyUsing the Snowpark APIHow it works...There’s more...See also
Working with Date and Time in Python
Technical requirementsWorking with DatetimeIndexHow to do it…How it works…There’s more…See alsoProviding a format argument to DateTimeHow to do it…How it works…There’s more…See alsoWorking with Unix epoch timestampsHow to do it…How it works…There’s more…See alsoWorking with time deltasHow to do it…How it works…There’s more…See alsoConverting DateTime with time zone informationHow to do it…How it works…There’s more…See alsoWorking with date offsetsHow to do it…How it works…There’s more…Using an existing holiday offsetDefining a custom holiday calendarSee alsoWorking with custom business daysHow to do it…How it works…There’s more…See alsoGet This Book’s PDF Version and Exclusive Extras
Handling Missing Data
Technical requirementsUnderstanding missing dataPerforming data quality checksGetting ready…How to do it…How it works…There’s more…See alsoHandling missing data with univariate imputation using pandasGetting readyHow to do it…How it works…There’s more…Using a Python dictionaryUsing another DataFrameSee alsoHandling missing data with univariate imputation using scikit-learnGetting readyHow to do it…How it works…There’s more…See alsoHandling missing data with multivariate imputationGetting readyHow to do it…Using IterativeImputerUsing KNNImputerHow it works…There’s more…See alsoHandling missing data with interpolationGetting readyHow to do it…How it works…There’s more…See also
Outlier Detection Using Statistical Methods
Technical requirementsThe nature of outliers in time-series dataResampling time-series dataHow to do it…How it works…There’s more...See alsoDetecting outliers using visualizationsGetting readyHow to do it...How it works...There’s more...See alsoDetecting outliers using the Tukey methodHow to do it...How it works...There’s more...See alsoDetecting outliers using a z-scoreHow to do it...How it works...There’s more...See alsoDetecting outliers using a modified z-scoreHow to do it...How it works...There’s more...See alsoDetecting outliers using the Hampel filterGetting readyHow to do it…How it works…There’s more…Detecting outliers with STRAYDetecting outliers with Matrix ProfileSee alsoGet This Book’s PDF Version and Exclusive Extras
Exploratory Data Analysis and Diagnosis
Technical requirementsDecomposing time series dataGetting readyHow to do it…Using seasonal_decomposeUsing STLHow it works…There’s more…See alsoDetecting time series stationarityGetting readyHow to do it…How it works…There’s more…See alsoApplying power transformationsGetting readyHow to do it…Testing normalityTesting homoskedasticityApplying a Box-Cox transformHow it works…There’s more...See alsoTesting for autocorrelationGetting readyHow to do it…How it works…There’s more...See also
Building Univariate Time Series Models Using Statistical Methods
Technical requirementsPlotting ACF and PACFHow to do it…How it works…There’s more...See alsoForecasting univariate time series data with exponential smoothingHow to do it…How it works…Simple exponential smoothingHolt’s exponential smoothingHolt-Winters’ exponential smoothingThere’s more…Using Darts exponential smoothingUsing AutoETSSee alsoForecasting univariate time series data with non-seasonal ARIMAGetting readyHow to do it…How it works…There’s more…See alsoForecasting univariate time series data with seasonal ARIMAHow to do it…How it works…There’s more…See alsoForecasting univariate time series with auto_arimaGetting readyHow to do it…How it works…There’s more…See alsoGet This Book’s PDF Version and Exclusive Extras

Additional Statistical Modeling Techniques for Time Series
Technical requirementsForecasting univariate time series data using ProphetGetting readyHow to do it…How it works…Trend configurationSeasonality configurationHoliday configurationUncertainty configurationThere’s more…Detecting changepointsPerforming cross-validationIncorporating holidays and regressorsSee alsoForecasting univariate time series data with ThetaGetting readyHow to do it…How it works…There’s more…See alsoForecasting multivariate time series data using VARGetting readyHow to do it…How it works…There’s more…See alsoEvaluating VAR modelsGetting readyHow to do it...How it works...There’s more...See alsoForecasting volatility in financial time series data with GARCHGetting readyHow to do it…How it works…There’s more...See alsoAutomated univariate forecasting using the StatsForecast libraryGetting readyHow to do it…How it works…There’s more…See also
Forecasting Using Supervised Machine Learning
Technical requirementsLoading the datasetsUnderstanding supervised MLPreparing time-series data for supervised learningHow to do it…How it works...There’s more...Window size selectionDate-based featuresStatistical featuresCyclical encodingExperimentationBest practicesSee alsoOne-step forecasting with scikit-learnHow to do it...How it works...There’s more...Feature selection and regularizationFeature importance in ensemble methodsSee alsoMulti-step forecasting with scikit-learnHow to do it...How it works...There’s more…See alsoForecasting with sktimeHow to do it...How it works...There’s more...Testing the multi-output strategyExploring the direct strategySee alsoHyperparameter tuning with sktimeGetting readyHow to do it...How it works...There’s more...Option 1: Using the MeanAbsoluteError classOption 2: Using the mean_absolute_error functionSee alsoForecasting with exogenous variables and ensemble learning in sktimeGetting readyHow to do it...How it works...There’s more...See alsoForecasting with XGBoostGetting readyHow to do it…How it works…There’s more…See alsoGet This Book’s PDF Version and Exclusive Extras
Deep Learning for Time Series Forecasting
Technical requirementsUnderstanding artificial neural networksForecasting with LSTM (Keras, Darts, and NeuralForecast)Getting readyHow to do it...Using Keras: one-step forecastingUsing the Darts RNNModel class (one-step forecast)Using Darts RNNModel for recursive multi-outputUsing Darts BlockRNNModel for direct multi-outputUsing NeuralForecast for direct multi-outputHow it works...There’s more...See alsoHandling multivariate time series and exogenous variablesHow to do it…Using BlockRNNModel from DartsUsing LSTM from NeuralForecastHow it works…There’s more…See alsoHyperparameter tuning for deep learning forecastersGetting readyHow to do it…Using Optuna with DartsUsing Optuna with NeuralForecastHow it works…There’s more…See alsoForecasting with TCNsGetting readyHow to do it…How it works…There’s more…See alsoForecasting with Transformers (N-HiTS)Getting readyHow to do it…How it works…There’s more…See also
Outlier Detection Using Unsupervised Machine Learning
Technical requirementsUnderstanding considerations for time series outlier detectionDetecting outliers using distance-based algorithmsKNN for global outlier detectionLOF for local outlier detectionHow to do it...Detecting point outliersDetecting outliers after decompositionDetecting contextual outliers with a sliding windowDetecting contextual outliers with feature engineeringDetecting contextual outliers with PyODAdapterHow it works...There’s more...See alsoDetecting outliers using clustering-based algorithmsCBLOF perspectiveHow to do it…How it works…There’s more…See alsoDetecting outliers using probabilistic and statistical algorithmsHow to do it...How it works...There’s more...See alsoDetecting outliers using kernel-based algorithms (PyOD)How to do it...How it works...There’s more...See alsoDetecting outliers using ensemble method algorithmsHow to do it...How it works...There’s more...See alsoDetecting outliers using deep learning-based algorithms (AutoEncoder)Getting readyHow to do it…How it works...There’s more…See alsoGet This Book’s PDF Version and Exclusive Extras
Advanced Techniques for Complex Time Series
Technical requirementsUnderstanding state-space modelsDecomposing time series with multiple seasonal patterns using MSTLGetting readyHow to do it...How it works...There’s more...See alsoForecasting with multiple seasonal patterns using the Unobserved Components Model (UCM)How to do it...How it works...There’s more...See alsoForecasting time series with multiple seasonal patterns using AutoTBATSGetting readyHow to do it…How it works…There’s more…See alsoForecasting time series with multiple seasonal patterns using NeuralProphetGetting readyHow to do it...How it works...There’s more...See also
Unlock Your Exclusive Benefits
Unlock this Book’s Free Benefits in 3 Easy Steps
Other Books You May Enjoy
Index

Content preview from Time Series Analysis with Python Cookbook - Second Edition

10 Additional Statistical Modeling Techniques for Time Series

In Chapter 9, you were introduced to popular forecasting techniques such as exponential smoothing, non-seasonal ARIMA, and seasonal ARIMA. These classical statistical approaches are widely used due to their speed, simplicity, and interpretability. Building on that foundation, this chapter introduces advanced statistical methods and powerful libraries that automate time series forecasting and model optimization.

You will learn about Prophet (developed by Facebook, now Meta), a robust tool for handling seasonality and holiday effects in time series forecasting. We’ll also look into statsmodels’ vector autoregressive (VAR) class for multivariate time series modeling and the arch library, ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Time Series Analysis with Python Cookbook

Publisher Resources

ISBN: 9781805124283

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Time Series Analysis with Python Cookbook - Second Edition

by Tarek A. Atwan

10

Additional Statistical Modeling Techniques for Time Series

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.