Table of Contents

• 1. Introduction

• 2. Importing data in colab notebook

  • 2.1 Importing data from local storage

  • 2.2 Importing data from Google Drive

  • 2.3 Importing data from a remote URL

  • 2.4 Importing data from a GitHub repository

  • 2.5 Google Cloud Storage (GCS)

• 3. Data Wrangling

  • 3.1 handling missing values

  • 3.2 Dealing with outliers

  • 3.3 Resampling and aggregation

  • 3.4 Handling inconsistent formats

  • 3.5 Feature engineering

  • 3.6 Lag Plots

  • 3.7 Normalization and scaling

• 4. Time Series Concepts

  • 4.1 What is a Time Series?

  • 4.2 Applications of Time Series Analysis

  • 4.3 Types of Time Series Data

  • 4.4 Time Series Components

  • 4.5 Time Series Decomposition

  • 4.6 Stationarity and Non-stationarity

• 5. Exploratory Data Analysis (EDA) for Time Series

  • 5.1 Visualizing Time Series Data

  • 5.2 Identifying Trends and Seasonality

    • 5.2.1 Additive and Multiplicative Decomposition

    • 5.2.2. Moving Averages(Smoothing)

    • 5.2.3. STL Decomposition (Seasonal-Trend decomposition using LOESS)

  • 5.3 ACF and PACF plots

• 6. Data Prepration

  • 6.1 Data cleaning

  • 6.2 Time alignment

  • 6.3 Resampling

  • 6.4 Smoothing

  • 6.5 Differencing

  • 6.6 Normalization

  • 6.7 Feature engineering

• 7. Stationarity

  • 7.1 How to make a time series stationary

    • 7.1.1 Differencing and its Importance

      • Example Code

    • 7.1.2 Seasonal Decomposition

• 8. Discovered a suite of classical time series forecasting methods

  • 8.1 Time Series Forecasting

  • 8.2 Concerns of Forecasting

  • 8.3 Examples of Time Series Forecasting

  • 8.4 Classical time series forecasting methods

• 9. Metrics

  • 9.1 Mean Absolute Error (MAE)

  • 9.2 The mean error (ME)

  • 9.3 Root Mean Squared Error (RMSE)

  • 9.4 Mean Absolute Percentage Error (MAPE)

  • 9.5 Symmetric Mean Absolute Percentage Error (SMAPE)

  • 9.6 Theil’s U-Statistic

  • 9.7 MRAE

• 10. Classical models

  • 10.1 Autoregression (AR)

  • 10.2 Moving Average (MA)

  • 10.3 Autoregressive Moving Average (ARMA)

  • 10.4 Autoregressive Integrated Moving Average (ARIMA)

  • 10.5 Seasonal Autoregressive Integrated Moving-Average (SARIMA)

  • 10.6 Seasonal Autoregressive Integrated Moving-Average with Exogenous Regressors (SARIMAX)

  • 10.7 Vector Autoregression (VAR)

  • 10.8 Vector Autoregression Moving-Average (VARMA)

  • 10.9 Vector Autoregression Moving-Average with Exogenous Regressors (VARMAX)

  • 10.10 Simple Exponential Smoothing (SES)

  • 10.11 Holt Winter’s Exponential Smoothing (HWES)

• 11. Deep Learning for Time Series Forecasting

• 12. Model selection

  • 12.1 Recurrent Neural Networks (RNNs)

    • 12.1.1 Single layer LSTM

    • 12.1.2 Stacked LSTM

    • 12.1.3 Bidirectional LSTM

    • 12.1.4 Encoder-Decoder LSTM

    • 12.1.5 Attention-based LSTM

    • 12.1.6 Hybrid RNN model

  • 12.2 Convolutional Neural Networks (CNNs)

    • 12.2.1 1D CNN

    • 12.2.2 Dilated CNN

    • 12.2.3 Temporal Convolutional Network (TCN)

    • 12.2.4 ConvLSTM

    • 12.2.5 Hybrid CNN models

  • 12.3 Transformer Models

    • 12.3.1 Transformer for Time Series Analysis

    • 12.3.2 Implementation Details for TSA

    • 12.3.3 Challenges and Considerations

    • 12.3.4 Recent Developments

    • Conclusion

  • 12.4 Autoencoders

    • 12.4.1 What is an Autoencoder?

    • 12.4.2 Autoencoders in Time Series Analysis

    • 12.4.3 Implementation Details for TSA

    • 12.4.4 Challenges and Considerations

    • 12.4.5 Conclusion

  • 12.5 Generative Adversarial Networks (GANs)

    • 12.5.1 Basics of GAN

    • 12.5.2 GANs in Time Series Analysis

    • 12.5.3 Implementation Details for TSA

    • 12.5.4 Challenges and Considerations

    • 12.5.5 Conclusion

• 13. Feature Engineering

• 14. Preprocessing using Deep Learning

  • 14.1 Handling missing values using an RNN

  • 14.2 Outlier detection using an autoencoder

  • 14.3 Handling seasonality and trends using a CNN

• 15. Time Series Analysis toolkits

  • 15.1 Scikit-learn

    • 15.1.1 How to use

    • 15.1.2 Example Code

  • 15.2. Statsmodels

    • 15.2.1 How to use

    • 15.2.2 Example Code

  • 15.3 Pandas

    • 15.3.1 How to use

    • 15.3.2 Example Code

  • 15.4 Prophet

    • 15.4.1 How to use

    • 15.4.2 Example Code

  • 15.5. sktime

    • 15.5.1 How to use

    • 15.5.2 Example Code

  • 15.6 Tslearn

    • 15.6.1 How to use

    • 15.6.2 Example Code

  • 15.7 Darts

    • 15.7.1 How to use

    • 15.7.2 Example Code

  • 15.8 PyFlux

    • 15.8.1 How to use

    • 15.8.2 Example Code

  • 15.9 SFRESH

    • 15.9.1 How to use

    • 15.9.2 2.Example Code

  • 15.11 Pastas

    • 15.11.1 How to use

    • 15.11.2 Example Code

• 16. NeuralProphet

  • 16.1 Installation

  • 16.2 Basic Usage

    • 16.2.1 Importing Libraries

    • 16.2.2 Generating Synthetic Data

    • 16.2.3 Model Training

    • 16.2.4 Forecasting

    • 16.2.5 Manualy Visualization of Forecast

    • 16.2.6 Visualizing Components

    • 16.2.7 Visualizing change points

    • 16.2.8 Manual Visualization of Change Points

    • 16.2.9 Plot using plotly backend