Become proficient in deriving insights from time-series data and analyzing a model’s performance
Key Features
• Explore popular and modern machine learning methods including the latest online and deep learning algorithms
• Learn to increase the accuracy of your predictions by matching the right model with the right problem
• Master time-series via real-world case studies on operations management, digital marketing, finance, and healthcare
Book Description
Machine learning has emerged as a powerful tool to understand hidden complexities in time-series datasets, which frequently need to be analyzed in areas as diverse as healthcare, economics, digital marketing, and social sciences. These datasets are essential for forecasting and predicting outcomes or for detecting anomalies to support informed decision making.
This book covers Python basics for time-series and builds your understanding of traditional autoregressive models as well as modern non-parametric models. You will become confident with loading time-series datasets from any source, deep learning models like recurrent neural networks and causal convolutional network models, and gradient boosting with feature engineering.
Machine Learning for Time-Series with Python explains the theory behind several useful models and guides you in matching the right model to the right problem. The book also includes real-world case studies covering weather, traffic, biking, and stock market data.
By the end of this book, you will be proficient in effectively analyzing time-series datasets with machine learning principles.
What you will learn
• Understand the main classes of time-series and learn how to detect outliers and patterns
• Choose the right method to solve time-series problems
• Characterize seasonal and correlation patterns through autocorrelation and statistical techniques
• Get to grips with time-series data visualization
• Understand classical time-series models like ARMA and ARIMA
• Implement deep learning models like Gaussian processes and transformers and state-of-the-art machine learning models
• Become familiar with many libraries like prophet, xgboost, and TensorFlow
Who This Book Is For
This book is ideal for data analysts, data scientists, and Python developers who are looking to perform time-series analysis to effectively predict outcomes. Basic knowledge of the Python language is essential. Familiarity with statistics is desirable.
Author(s): Ben Auffarth
Edition: 1
Publisher: Packt Publishing - ebooks Account
Year: 2021
Language: English
Commentary: Vector PDF
Pages: 371
City: Birmingham, UK
Tags: Machine Learning; Deep Learning; Unsupervised Learning; Reinforcement Learning; Supervised Learning; Python; Recurrent Neural Networks; Autoencoders; Bayesian Inference; NumPy; pandas; Jupyter; Autoregression; Forecasting; Prophet; Time Series Analysis; Markov Models; Data Preprocessing; Moving Average
Cover
Copyright
Contributors
Table of Contents
Preface
Chapter 1: Introduction to Time Series with Python
What Is a Time Series?
Characteristics of Time Series
Time Series and Forecasting – Past and Present
Demography
Genetics
Astronomy
Economics
Meteorology
Medicine
Applied Statistics
Python for Time Series
Installing libraries
Jupyter Notebook and JupyterLab
NumPy
pandas
Best practice in Python
Summary
Chapter 2: Time-Series Analysis with Python
What is time series analysis?
Working with time series in Python
Requirements
Datetime
pandas
Understanding the variables
Uncovering relationships between variables
Identifying trend and seasonality
Summary
Chapter 3: Preprocessing Time Series
What Is Preprocessing?
Feature Transforms
Scaling
Log and Power Transformations
Imputation
Feature Engineering
Date- and Time-Related Features
ROCKET
Shapelets
Python Practice
Log and Power Transformations in Practice
Imputation
Holiday Features
Date Annotation
Paydays
Seasons
The Sun and Moon
Business Days
Automated Feature Extraction
ROCKET
Shapelets in Practice
Summary
Chapter 4: Introduction to Machine Learning for Time-Series
Machine learning with time series
Supervised, unsupervised, and reinforcement learning
History of machine learning
Machine learning workflow
Cross-validation
Error metrics for time series
Regression
Classification
Comparing time-series
Machine learning algorithms for time-series
Distance-based approaches
Shapelets
ROCKET
Time Series Forest and Canonical Interval Forest
Symbolic approaches
HIVE-COTE
Discussion
Implementations
Summary
Chapter 5: Time-Series Forecasting with Moving Averages and Autoregressive Models
What are classical models?
Moving average and autoregression
Model selection and order
Exponential smoothing
ARCH and GARCH
Vector autoregression
Python libraries
Statsmodels
Python practice
Requirements
Modeling in Python
Summary
Chapter 6: Unsupervised Methods for Time-Series
Unsupervised methods for time-series
Anomaly detection
Microsoft
Google
Amazon
Facebook
Twitter
Implementations
Change point detection
Clustering
Python practice
Requirements
Anomaly detection
Change point detection
Summary
Chapter 7: Machine Learning Models for Time-Series
More machine learning methods for time series
Validation
K-nearest neighbors with dynamic time warping
Silverkite
Gradient boosting
Python exercise
Virtual environments
K-nearest neighbors with dynamic time warping in Python
Silverkite
Gradient boosting
Ensembles with Kats
Summary
Chapter 8: Online Learning for Time-Series
Online learning for time series
Online algorithms
Drift
Drift detection methods
Adaptive learning methods
Python practice
Drift detection
Regression
Model selection
Summary
Chapter 9: Probabilistic Models for Time-Series
Probabilistic Models for Time-Series
Prophet
Markov Models
Fuzzy Modeling
Bayesian Structural Time-Series Models
Python Exercise
Prophet
Markov Switching Model
Fuzzy Time-Series
Bayesian Structural Time-Series Modeling
Summary
Chapter 10: Deep Learning for Time-Series
Introduction to deep learning
Deep learning for time series
Autoencoders
InceptionTime
DeepAR
N-BEATS
Recurrent neural networks
ConvNets
Transformer architectures
Informer
Python practice
Fully connected network
Recurrent neural network
Dilated causal convolutional neural network
Summary
Chapter 11: Reinforcement Learning for Time-Series
Introduction to reinforcement learning
Reinforcement Learning for Time-Series
Bandit algorithms
Deep Q-Learning
Python Practice
Recommendations
Trading with DQN
Summary
Chapter 12: Multivariate Forecasting
Forecasting a Multivariate Time-Series
Python practice
What's next for time-series?
Why subscribe?
Packt Page
Other Books You May Enjoy
Index
