A Practical Approach to Timeseries Forecasting Using Python

A Practical Approach to Timeseries Forecasting Using Python

by AI Sciences
Released March 2023
Publisher(s): Packt Publishing
ISBN: 9781837632510

Watch it now on the O’Reilly learning platform with a 10-day free trial.

O’Reilly members get unlimited access to books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.

Start your free trial

Video description

Have you ever wondered how weather predictions, population estimates, and even the lifespan of the universe are made?

Discover the power of time series forecasting with state-of-the-art ML and DL models.

The course begins with the fundamentals of time series analysis, including its characteristics, applications in real-world scenarios, and practical examples. Then progress to exploring data analysis and visualization techniques for time series data, ranging from basic to advanced levels, using powerful libraries such as NumPy, Pandas, and Matplotlib. Python will be utilized to assess various aspects of your time series data, such as seasonality, trend, noise, autocorrelation, mean over time, correlation, and stationarity.

Additionally, you will learn how to pre-process time series data for utilization in applied machine learning and recurrent neural network models, which will enable you to train, test, and assess your forecasted results. Finally, you will acquire an understanding of the applied ML models, including their performance evaluations and comparisons.

In the RNNs module, you will be building GRU, LSTM, Stacked LSTM, BiLSTM, and Stacked BiLSTM models.

By the end of this course, you will be able to understand time series forecasting and its parameters, evaluate the ML models, and evaluate the model and implement RNN models for time series forecasting.

What You Will Learn

  • Learn data analysis techniques and handle time series forecasting
  • Implement data visualization techniques using Matplotlib
  • Evaluate applied machine learning in time series forecasting
  • Look at auto regression, ARIMA, Auto ARIMA, SARIMA, and SARIMAX
  • Learn to model LSTM, Stacked LSTM, BiLSTM, and Stacked BiLSTM models
  • Implement ML and RNN models with three state-of-the-art projects

Audience

No prior knowledge of DL, data analysis, or math is required. You will start from the basics and gradually build your knowledge of the subject. Only the basics of Python are required.

This course is designed for both beginners with some programming experience and even those who know nothing about data analysis, ML, and RNNs.

The course is suitable for individuals who want to advance their skills in ML and DL, master the relation of data science with time series analysis, implement time series parameters and evaluate their impact on it and implement ML algorithms for time series forecasting.

About The Author

AI Sciences: AI Sciences are experts, PhDs, and artificial intelligence practitioners, including computer science, machine learning, and Statistics. Some work in big companies such as Amazon, Google, Facebook, Microsoft, KPMG, BCG, and IBM.

AI sciences produce a series of courses dedicated to beginners and newcomers on techniques and methods of machine learning, statistics, artificial intelligence, and data science. They aim to help those who wish to understand techniques more easily and start with less theory and less extended reading. Today, they publish more comprehensive courses on specific topics for wider audiences.

Their courses have successfully helped more than 100,000 students master AI and data science.

Table of contents

  1. Chapter 1 : Introduction
    1. Introduction to Time Series Forecast
    2. Introduction to Instructor
    3. Course Introduction
  2. Chapter 2 : Motivation and Overview of Time Series Analysis
    1. Introduction to Time Series Forecasting
    2. Features of Time Series
    3. Types of Time Series Data
    4. Stages for Time Series Forecasting
    5. Data Manipulation in Time Series
    6. Data Processing for Time Series Forecasting
    7. Machine Learning Forecasting
    8. RNN Forecasting
    9. Projects to Be Covered
  3. Chapter 3 : Basics of Data Manipulation in Time Series
    1. Module Overview
    2. Packages Required to Execute Codes Error-Free
    3. Overview of Basic Plotting and Visualization
    4. Overview of Time Series Parameters
    5. Dependencies Installation and Dataset Overview
    6. Data Manipulation in Python
    7. Data Slicing and Indexing
    8. Basic Data Visualization with Single Time Series Feature
    9. Data Visualization with Multiple Time Series Features
    10. Data Visualization with Customized Features Selection
    11. Area Plots in Data Analysis
    12. Histogram with Single Feature
    13. Histogram Multiple Features
    14. Pie Charts
    15. Time Series Parameters
    16. Quiz Video
    17. Quiz Solution
  4. Chapter 4 : Data Processing for Timeseries Forecasting
    1. Module Overview
    2. Dataset Significance
    3. Dataset Overview
    4. Dataset Manipulation
    5. Data Pre-Processing
    6. RVT Models
    7. Automatic Time Series Decomposition
    8. Trend Using Moving Average Filter
    9. Seasonality Comparison
    10. Resampling
    11. Noise in Time Series
    12. Feature Engineering
    13. Stationarity in Time Series
    14. Handling Non-Stationarity in Time Series
    15. Quiz
    16. Quiz Solution
  5. Chapter 5 : Machine Learning in Time Series Forecasting
    1. Section Overview
    2. Data Preparation
    3. Auto Correlation and Partial Correlation
    4. Data Splitting
    5. Autoregression
    6. Autoregression in Python
    7. Moving Average and ARMA
    8. ARIMA
    9. ARIMA in Python
    10. Auto ARIMA in Python
    11. SARIMA
    12. SARIMA in Python
    13. Auto SARIMA in Python
    14. Future Predictions Using SARIMA
    15. Quiz
    16. Quiz Solution
  6. Chapter 6 : Recurrent Neural Networks in Time Series Forecasting
    1. Module Overview
    2. Important Parameters
    3. LSTM Models
    4. BiLSTM Models
    5. GRU Models
    6. Underfitting and Overfitting
    7. Model for Underfitting and Overfitting
    8. Model Evaluation for Underfitting and Overfitting
    9. Dataset Preparation and Scaling
    10. Dataset Reshaping
    11. LSTM Implementation on Dataset
    12. Time Series Forecasting (TSF) Using LSTM
    13. Graph for TSF Using LSTM
    14. LSTM Parameter Change and Stacked LSTM
    15. BiLSTM for Time Series Forecasting
    16. Quiz
    17. Quiz Solution
  7. Chapter 7 : Project 1: COVID-19 Positive Cases Prediction Using Machine Learning Algorithm
    1. Project Overview
    2. Dataset Overview
    3. Dataset Correlation
    4. Shape and NULL Check
    5. Dataset Index
    6. Visualize the Data
    7. Area Plot
    8. Autocorrelation, Standard Deviation, and Mean
    9. Stationarity Check
    10. ARIMA Implementation
    11. SARIMA Implementation
    12. Variations in SARIMA
  8. Chapter 8 : Project 2: Microsoft Corporation Stock Prediction Using RNNs
    1. Module Overview
    2. Data Analysis
    3. Data Visualization Line Plots
    4. Area Plots
    5. Auto Correlation, Standard Deviation, and Mean
    6. Stationarity Check
    7. Data Manipulation for Deep Learning
    8. Dataset Division
    9. LSTM Implementation and Errors
    10. LSTM Forecasting
    11. Stacked LSTM Forecasting
    12. BiLSTM and Stacked BiLSTM
  9. Chapter 9 : Project 3: Birth Rate Forecasting Using RNNs with Advanced Data Analysis
    1. Project Overview
    2. Dataset Overview
    3. Yearly Birth Distribution Plot and Birth Rate Plot
    4. Monthly Birth Distribution Plot and Birth Rate Plot
    5. Day-Wise and Date-Wise Birth Distribution Plot and Birth Rate Plot
    6. Birth Rate Range Plot
    7. Data Manipulation
    8. Stationarity Check
    9. Manipulation for Forecasting
    10. Scaling
    11. LSTM Forecasting
    12. Stacked LSTM and BiLSTM
    13. Course Conclusion

Product information

  • Title: A Practical Approach to Timeseries Forecasting Using Python
  • Author(s): AI Sciences
  • Release date: March 2023
  • Publisher(s): Packt Publishing
  • ISBN: 9781837632510