Book Image

Time Series Analysis with Python Cookbook

By : Tarek A. Atwan
Book Image

Time Series Analysis with Python Cookbook

By: Tarek A. Atwan

Overview of this book

Time series data is everywhere, available at a high frequency and volume. It is complex and can contain noise, irregularities, and multiple patterns, making it crucial to be well-versed with the techniques covered in this book for data preparation, analysis, and forecasting. This book covers practical techniques for working with time series data, starting with ingesting time series data from various sources and formats, whether in private cloud storage, relational databases, non-relational databases, or specialized time series databases such as InfluxDB. Next, you’ll learn strategies for handling missing data, dealing with time zones and custom business days, and detecting anomalies using intuitive statistical methods, followed by more advanced unsupervised ML models. The book will also explore forecasting using classical statistical models such as Holt-Winters, SARIMA, and VAR. The recipes will present practical techniques for handling non-stationary data, using power transforms, ACF and PACF plots, and decomposing time series data with multiple seasonal patterns. Later, you’ll work with ML and DL models using TensorFlow and PyTorch. Finally, you’ll learn how to evaluate, compare, optimize models, and more using the recipes covered in the book.

Related Content you might be interested in

Current Title:

Small book image
Time Series Analysis with Python Cookbook
Tarek A. Atwan
Jun, 2022 | 630 pages
Small book image
Practical Time Series Analysis
Avishek Pal | PKS Prakash
Sep, 2017 | 244 pages
Small book image
Forecasting Time Series Data with Facebook Prophet
Greg Rafferty
Mar, 2021 | 270 pages
Small book image
Codeless Time Series Analysis with KNIME
KNIME AG | Maarit Widmann | Corey Weisinger | Daniele Tonini
Aug, 2022 | 392 pages
Table of Contents (18 chapters)
Preface
Preface
Who this book is for
What this book covers
To get the most out of this book
Download the example code files
Code in Action
Conventions used
Sections
Get in touch
Share Your Thoughts
1
Chapter 1: Getting Started with Time Series Analysis
Chapter 1: Getting Started with Time Series Analysis
Technical requirements
Development environment setup
Installing Python libraries
Installing JupyterLab and JupyterLab extensions
Free Chapter
2
Chapter 2: Reading Time Series Data from Files
Chapter 2: Reading Time Series Data from Files
Technical requirements
Reading data from CSVs and other delimited files
Reading data from an Excel file
Reading data from URLs
Reading data from a SAS dataset
3
Chapter 3: Reading Time Series Data from Databases
Chapter 3: Reading Time Series Data from Databases
Technical requirements
Reading data from a relational database
Reading data from Snowflake
Reading data from a document database (MongoDB)
Reading third-party financial data using APIs
Reading data from a time series database (InfluxDB)
4
Chapter 4: Persisting Time Series Data to Files
Chapter 4: Persisting Time Series Data to Files
Technical requirements
Serializing time series data with pickle
Writing to CSV and other delimited files
Writing data to an Excel file
Storing data to S3
5
Chapter 5: Persisting Time Series Data to Databases
Chapter 5: Persisting Time Series Data to Databases
Technical requirements
Writing time series data to a relational database (PostgreSQL and MySQL)
Writing time series data to MongoDB
Writing time series data to InfluxDB
Writing time series data to Snowflake
6
Chapter 6: Working with Date and Time in Python
Chapter 6: Working with Date and Time in Python
Technical requirements
Working with DatetimeIndex
Providing a format argument to DateTime
Working with Unix epoch timestamps
Working with time deltas
Converting DateTime with time zone information
Working with date offsets
Working with custom business days
7
Chapter 7: Handling Missing Data
Chapter 7: Handling Missing Data
Technical requirements
Understanding missing data
Performing data quality checks
Handling missing data with univariate imputation using pandas
Handling missing data with univariate imputation using scikit-learn
Handling missing data with multivariate imputation
Handling missing data with interpolation
8
Chapter 8: Outlier Detection Using Statistical Methods
Chapter 8: Outlier Detection Using Statistical Methods
Technical requirements
Understanding outliers
Resampling time series data
Detecting outliers using visualizations
Detecting outliers using the Tukey method
Detecting outliers using a z-score
Detecting outliers using a modified z-score
9
Chapter 9: Exploratory Data Analysis and Diagnosis
Chapter 9: Exploratory Data Analysis and Diagnosis
Technical requirements
Plotting time series data using pandas
Plotting time series data with interactive visualizations using hvPlot
Decomposing time series data
Detecting time series stationarity
Applying power transformations
Testing for autocorrelation in time series data
10
Chapter 10: Building Univariate Time Series Models Using Statistical Methods
Chapter 10: Building Univariate Time Series Models Using Statistical Methods
Technical requirements
Plotting ACF and PACF
Forecasting univariate time series data with exponential smoothing
Forecasting univariate time series data with non-seasonal ARIMA
Forecasting univariate time series data with seasonal ARIMA
11
Chapter 11: Additional Statistical Modeling Techniques for Time Series
Chapter 11: Additional Statistical Modeling Techniques for Time Series
Technical requirements
Forecasting time series data using auto_arima
Forecasting time series data using Facebook Prophet
Forecasting multivariate time series data using VAR
Evaluating vector autoregressive (VAR) models
Forecasting volatility in financial time series data with GARCH
12
Chapter 12: Forecasting Using Supervised Machine Learning
Chapter 12: Forecasting Using Supervised Machine Learning
Technical requirements
Understanding supervised machine learning
Preparing time series data for supervised learning
One-step forecasting using linear regression models with scikit-learn
Multi-step forecasting using linear regression models with scikit-learn
Forecasting using non-linear models with sktime
Optimizing a forecasting model with hyperparameter tuning
Forecasting with exogenous variables and ensemble learning
13
Chapter 13: Deep Learning for Time Series Forecasting
Chapter 13: Deep Learning for Time Series Forecasting
Technical requirements
Understanding artificial neural networks
Forecasting with an RNN using Keras
Forecasting with LSTM using Keras
Forecasting with a GRU using Keras
Forecasting with an RNN using PyTorch
Forecasting with LSTM using PyTorch
Forecasting with a GRU using PyTorch
14
Chapter 14: Outlier Detection Using Unsupervised Machine Learning
Chapter 14: Outlier Detection Using Unsupervised Machine Learning
Technical requirements
Detecting outliers using KNN
Detecting outliers using LOF
Detecting outliers using iForest
Detecting outliers using One-Class Support Vector Machine (OCSVM)
Detecting outliers using COPOD
Detecting outliers with PyCaret
15
Chapter 15: Advanced Techniques for Complex Time Series
Chapter 15: Advanced Techniques for Complex Time Series
Technical requirements
Understanding state-space models
Decomposing time series with multiple seasonal patterns using MSTL
Forecasting with multiple seasonal patterns using the Unobserved Components Model (UCM)
Forecasting time series with multiple seasonal patterns using Prophet
Forecasting time series with multiple seasonal patterns using NeuralProphet
16
Index
Index
Why subscribe?
17
Other Books You May Enjoy
Other Books You May Enjoy
Packt is searching for authors like you
Share Your Thoughts

Reading Time Series Data from Files

In this chapter, we will use pandas, a popular Python library with a rich set of I/O tools, data wrangling, and date/time functionality to streamline working with time series data. In addition, you will explore several reader functions available in pandas to ingest data from different file types, such as Comma-Separated Value (CSV), Excel, and SAS. You will explore reading from files, whether they are stored locally on your drive or remotely on the cloud, such as an AWS S3 bucket.

Time series data is complex and can be in different shapes and formats. Conveniently, the pandas reader functions offer a vast number of arguments (parameters) to help handle such variety in the data.

The pandas library provides two fundamental data structures, Series and DataFrame, implemented as classes. The DataFrame class is a distinct data structure for working with tabular data (think rows and columns in a spreadsheet). The main difference between the two data structures is that a Series is one-dimensional (single column), and a DataFrame is two-dimensional (multiple columns). The relationship between the two is that you get a Series when you slice out a column from a DataFrame. You can think of a DataFrame as a side-by-side concatenation of two or more Series objects.

A particular feature of the Series and DataFrames data structures is that they both have a labeled axis called index. A specific type of index that you will often see with time series data is the DatetimeIndex which you will explore further in this chapter. Generally, the index makes slicing and dicing operations very intuitive. For example, to make a DataFrame ready for time series analysis, you will learn how to create DataFrames with an index of type DatetimeIndex.

We will cover the following recipes on how to ingest data into a pandas DataFrame:

  • Reading data from CSVs and other delimited files
  • Reading data from an Excel file
  • Reading data from URLs
  • Reading data from a SAS dataset

    Why DatetimeIndex?

    A pandas DataFrame with an index of type DatetimeIndex unlocks a large set of features and useful functions needed when working with time series data. You can think of it as adding a layer of intelligence or awareness to pandas to treat the DataFrame as a time series DataFrame.

Previous Section
End of Section 1
Next Section