Hands-on Time Series Analysis with Python : From Basics to Bleeding Edge Techniques

Learn the concepts of time series from traditional to leading-edge techniques. This book uses comprehensive examples to clearly illustrate statistical approaches and methods of analyzing time series data and its utilization in the real world. All the code is available in Jupyter notebooks. You'll begin by reviewing time series fundamentals, the structure of time series data, pre-processing, and how to craft the features through data wrangling. Next, you'll look at traditional time series techniques like ARMA, SARIMAX, VAR, and VARMA using trending framework like StatsModels and pmdarima. The book also explains building classification models using sktime, and covers advanced deep learning-based techniques like ANN, CNN, RNN, LSTM, GRU and Autoencoder to solve time series problem using Tensorflow. It concludes by explaining the popular framework fbprophet for modeling time series analysis. After reading __Hands -On Time Series Analysis with Python__ , you'll be able to apply these new techniques in industries, such as oil and gas, robotics, manufacturing, government, banking, retail, healthcare, and more. **What You'll Learn** \* \* Explains basics to advanced concepts of time series. \* How to design, develop, train, test and validate time-series methodologies. \* What are Smoothing, ARMA, ARIMA, SARIMA,SRIMAX, VAR, VARMA techniques in time series and how to optimally tune parameters to yield best results. \* Learn how to leverage bleeding-edge techniques such as ANN, CNN, RNN, LSTM, GRU, Autoencoder to solve both Univariate and multivariate problems by using two types of data prepration methods for time series. \* Univariate and multivariate problem solving using fbprophet. **Who This Book Is For** Data scientists, data analysts, financial analysts, and stock market researchers \*\* Table of Contents......Page 4 About the Authors......Page 11 About the Technical Reviewer......Page 13 Acknowledgments......Page 14 Introduction......Page 15 Chapter 1: Time-Series Characteristics......Page 16 Time-Series Data......Page 17 Panel Data/Longitudinal Data......Page 19 Detecting Trend Using a Hodrick-Prescott Filter......Page 21 Detrending a Time Series......Page 22 Detrending Using Pandas Differencing......Page 23 Detrending Using a SciPy Signal......Page 24 Detrend Using an HP Filter......Page 25 Seasonality......Page 26 Multiple Box Plots......Page 27 Autocorrelation Plot......Page 28 Deseasoning of Time-Series Data......Page 29 Seasonal Decomposition......Page 30 Cyclic Variations......Page 31 Detecting Cyclical Variations......Page 32 Decomposing a Time Series into Its Components......Page 33 Summary......Page 36 Chapter 2: Data Wrangling and Preparation for Time Series......Page 37 Loading Data Using CSV......Page 38 Loading Data Using JSON......Page 39 Loading Data from a URL......Page 40 Selecting the Top Five Records......Page 41 Applying a Filter......Page 42 Distinct (Unique)......Page 43 IN......Page 44 NOT IN......Page 46 Ascending Data Order......Page 47 Descending Data Order......Page 48 Aggregation......Page 49 GROUP BY......Page 50 GROUP BY with Aggregation......Page 52 Join (Merge)......Page 53 INNER JOIN......Page 55 LEFT JOIN......Page 57 RIGHT JOIN......Page 60 OUTER JOIN......Page 62 Summary of the DataFrame......Page 65 Resampling......Page 66 Resampling by Year......Page 67 Resampling by Week......Page 68 Windowing Function......Page 69 Rolling Window......Page 70 Exponentially Weighted Moving Window......Page 71 Shifting......Page 72 Handling Missing Data......Page 74 FFILL......Page 76 FILLNA......Page 77 Summary......Page 78 Chapter 3: Smoothing Methods......Page 79 Introduction to Simple Exponential Smoothing......Page 80 Simple Exponential Smoothing in Action......Page 82 Introduction to Double Exponential Smoothing......Page 90 Double Exponential Smoothing in Action......Page 92 Introduction to Triple Exponential Smoothing......Page 100 Triple Exponential Smoothing in Action......Page 101 Summary......Page 111 Types of Stationary Behavior in a Time Series......Page 112 Using Summary Statistics......Page 114 Using Statistics Unit Root Tests......Page 115 Augmented Dickey-Fuller Test......Page 117 Kwiatkowski-Phillips-Schmidt-Shin Test......Page 118 Differencing......Page 119 Random Walk......Page 120 First-Order Differencing (Trend Differencing)......Page 121 Second-Order Differencing (Trend Differencing)......Page 122 Second-Order Differencing for Seasonal Data......Page 123 Autoregressive Models......Page 124 Moving Average......Page 126 Autocorrelation and Partial Autocorrelation Functions......Page 129 Introduction to ARMA......Page 130 Autoregressive Model......Page 131 Introduction to Autoregressive Integrated Moving Average......Page 132 The Integration (I)......Page 134 ARIMA in Action......Page 135 Introduction to Seasonal ARIMA......Page 142 SARIMA in Action......Page 144 Introduction to SARIMAX......Page 156 SARIMAX in Action......Page 157 Introduction to Vector Autoregression......Page 167 VAR in Action......Page 168 VARMA in Action......Page 185 Summary......Page 197 Introduction to Neural Networks......Page 198 Perceptron......Page 199 Binary Step Function......Page 201 Linear Activation Function......Page 202 Nonlinear Activation Function......Page 203 Sigmoid......Page 204 Rectified Linear Unit......Page 205 Parametric ReLU......Page 206 Softmax......Page 207 Forward Propagation......Page 208 Backward Propagation......Page 209 Learning Rate vs. Gradient Descent Optimizers......Page 212 Recurrent Neural Networks......Page 215 Feed-Forward Recurrent Neural Network......Page 217 Backpropagation Through Time in RNN......Page 219 Long Short-Term Memory......Page 223 Step-by-Step Explanation of LSTM......Page 225 Peephole LSTM......Page 227 Peephole Convolutional LSTM......Page 228 Gated Recurrent Units......Page 229 Convolution Neural Networks......Page 232 Generalized CNN Formula......Page 235 One-Dimensional CNNs......Page 236 Auto-encoders......Page 237 Summary......Page 239 Single-Step Data Preparation for Time-Series Forecasting......Page 240 Horizon-Style Data Preparation for Time-Series Forecasting......Page 242 LSTM Univariate Single-Step Style in Action......Page 243 LSTM Univariate Horizon Style in Action......Page 255 Bidirectional LSTM Univariate Single-Step Style in Action......Page 266 Bidirectional LSTM Univariate Horizon Style in Action......Page 275 GRU Univariate Single-Step Style in Action......Page 284 GRU Univariate Horizon Style in Action......Page 292 Auto-encoder LSTM Univariate Single-Step Style in Action......Page 301 Auto-encoder LSTM Univariate Horizon Style in Action......Page 310 CNN Univariate Single-Step Style in Action......Page 319 CNN Univariate Horizon Style in Action......Page 328 Summary......Page 337 LSTM Multivariate Horizon Style in Action......Page 338 Bidirectional LSTM Multivariate Horizon Style in Action......Page 350 Auto-encoder LSTM Multivariate Horizon Style in Action......Page 359 GRU Multivariate Horizon Style in Action......Page 369 CNN Multivariate Horizon Style in Action......Page 378 Summary......Page 387 The Prophet Model......Page 388 Implementing Prophet......Page 389 Adding Log Transformation......Page 394 Adding Built-in Country Holidays......Page 399 Adding Exogenous variables using add_regressors(function)......Page 402 Summary......Page 407 Index......Page 408 Learn the concepts of time series from traditional to bleeding-edge techniques. This book uses comprehensive examples to clearly illustrate statistical approaches and methods of analyzing time series data and its utilization in the real world. All the code is available in Jupyter notebooks. You'll begin by reviewing time series fundamentals, the structure of time series data, pre-processing, and how to craft the features through data wrangling. Next, you'll look at traditional time series techniques like ARMA, SARIMAX, VAR, and VARMA using trending framework like StatsModels and pmdarima. The book also explains building classification models using sktime, and covers advanced deep learning-based techniques like ANN, CNN, RNN, LSTM, GRU and Autoencoder to solve time series problem using Tensorflow. It concludes by explaining the popular framework fbprophet for modeling time series analysis. After reading Hands -On Time Series Analysis with Python , you'll be able to apply these new techniques in industries, such as oil and gas, robotics, manufacturing, government, banking, retail, healthcare, and more. What You'll Learn: · Explains basics to advanced concepts of time series · How to design, develop, train, and validate time-series methodologies · What are smoothing, ARMA, ARIMA, SARIMA,SRIMAX, VAR, VARMA techniques in time series and how to optimally tune parameters to yield best results · Learn how to leverage bleeding-edge techniques such as ANN, CNN, RNN, LSTM, GRU, Autoencoder to solve both Univariate and multivariate problems by using two types of data preparation methods for time series. · Univariate and multivariate problem solving using fbprophet. Who This Book Is For Data scientists, data analysts, financial analysts, and stock market researchers This book explains the concepts of time series from traditional to bleeding-edge techniques with full-fledged examples. The book begins by covering time series fundamentals and its characteristics, the structure of time series data, pre-processing, and ways of crafting the features through data wrangling. Next, it covers the traditional time series techniques like Smoothing methods, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA using trending framework like StatsModels, pmdarima. Further, Book explains the building classification models using sktime, and covers how to leverage advance deep learning-based techniques like ANN, CNN, RNN, LSTM, GRU and Autoencoder to solve time series problem using Tensorflow. It finally concludes by explaining the popular framework fbprophet for modeling time series analysis. After completion of the book, the reader will have a good understanding of working with different techniques of time series methods. All the codes presented in this notebook are available in Jupyter notebooks, which allows readers to do hands-on and enhance them in exciting ways. What You'll Learn: Explains basics to advanced concepts of time series ; How to design, develop, train, and validate time-series methodologies ; What are smoothing, ARMA, ARIMA, SARIMA, SRIMAX, VAR, VARMA techniques in time series and how to optimally tune parameters to yield best results ; Learn how to leverage bleeding-edge techniques such as ANN, CNN, RNN, LSTM, GRU, Autoencoder to solve both Univariate and multivariate problems by using two types of data preparation methods for time series. Univariate and multivariate problem solving using fbprophet Chapter 1: Time Series and its Characteristics -- Chapter 2: Data Wrangling and Preparation for Time Series -- Chapter 3: Smoothing Methods -- Chapter 4: Regression Extension Techniques for Time Series -- Chapter 5: Bleeding Edge Techniques -- Chapter 6: Bleeding Edge Techniques for Univariate Time Series -- Chapter 7: Bleeding Edge Techniques for Multivariate Time Series -- Chapter 8: Prophet