Muutke küpsiste eelistusi

Time Series Forecasting in Python [Pehme köide]

4.31/5 (51 hinnangut Goodreads-ist)
  • Formaat: Paperback / softback, 456 pages, kõrgus x laius x paksus: 234x186x28 mm, kaal: 840 g
  • Ilmumisaeg: 10-Nov-2022
  • Kirjastus: Manning Publications
  • ISBN-10: 161729988X
  • ISBN-13: 9781617299889
Teised raamatud teemal:
Time Series Forecasting in PythonSuurem pilt
  • Formaat: Paperback / softback, 456 pages, kõrgus x laius x paksus: 234x186x28 mm, kaal: 840 g
  • Ilmumisaeg: 10-Nov-2022
  • Kirjastus: Manning Publications
  • ISBN-10: 161729988X
  • ISBN-13: 9781617299889
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781617299889.html
Märksõnad:
Build predictive models from time-based patterns in your data. Master statistical models including new deep learning approaches for time series forecasting.

In  Time Series Forecasting in Python  you will learn how to:





Recognize a time series forecasting problem and build a performant predictive model Create univariate forecasting models that account for seasonal effects and external variables Build multivariate forecasting models to predict many time series at once Leverage large datasets by using deep learning for forecasting time series Automate the forecasting process



DESCRIPTION  Time Series Forecasting in Python teaches you to build powerful predictive models from time-based data. Every model you create is relevant, useful, and easy to implement with Python. You'll explore interesting real-world datasets like Google's daily stock price and economic data for the USA, quickly progressing from the basics to developing large-scale models that use deep learning tools like TensorFlow.Time Series Forecasting in Python teaches you to apply time series forecasting and get immediate, meaningful predictions. You'll learn both traditional statistical and new deep learning models for time series forecasting, all fully illustrated with Python source code. Time Series Forecasting in Python teaches you to build powerful predictive models from time-based data. Every model you create is relevant, useful, and easy to implement with Python. You'll explore interesting real-world datasets like Google's daily stock price and economic data for the USA, quickly progressing from the basics to developing large-scale models that use deep learning tools like TensorFlow. about the technology Time series forecasting reveals hidden trends and makes predictions about the future from your data. This powerful technique has proven incredibly valuable across multiple fieldsfrom tracking business metrics, to healthcare and the sciences. Modern Python libraries and powerful deep learning tools have opened up new methods and utilities for making practical time series forecasts. about the book Time Series Forecasting in Python teaches you to apply time series forecasting and get immediate, meaningful predictions. You'll learn both traditional statistical and new deep learning models for time series forecasting, all fully illustrated with Python source code. Test your skills with hands-on projects for forecasting air travel, volume of drug prescriptions, and the earnings of Johnson & Johnson. By the time you're done, you'll be ready to build accurate and insightful forecasting models with tools from the Python ecosystem.
Preface xvii
Acknowledgments xix
About this book xx
About the author xxiv
About the cover illustration xxv
Part 1 Time waits for no one
1(58)
1 Understanding time series forecasting
3(11)
1.1 Introducing time series
4(4)
Components of a time series
5(3)
1.2 Bird's-eye view of time series forecasting
8(4)
Setting a goal
9(1)
Determining what must be forecast to achieve your goal
9(1)
Setting the horizon of the forecast
10(1)
Gathering the data
10(1)
Developing a forecasting model
10(1)
Deploying to production
11(1)
Monitoring
11(1)
Collecting new data
11(1)
1.3 How time series forecasting is different from other regression tasks
12(1)
Time series have an order
12(1)
Time series sometimes do not have features
13(1)
1.4 Next steps
13(1)
2 A naive prediction of the future
14(16)
2.1 Defining a baseline model
16(1)
2.2 Forecasting the historical mean
17(6)
Setup for baseline implementations
17(2)
Implementing the historical mean baseline
19(4)
2.3 Forecasting last year's mean
23(2)
2.4 Predicting using the last known value
25(1)
2.5 Implementing the naive seasonal forecast
26(2)
2.6 Next steps
28(2)
3 Going on a random walk
30(29)
3.1 The random walk process
31(4)
Simulating a random walk process
32(3)
3.2 Identifying a random walk
35(12)
Stationarity
36(2)
Testing for stationarity
38(3)
The autocorrelation function
41(1)
Putting it all together
42(3)
Is GOOGL a random walk?
45(2)
3.3 Forecasting a random walk
47(8)
Forecasting on a long horizon
48(4)
Forecasting the next timestep
52(3)
3.4 Next steps
55(1)
3.5 Exercises
56(3)
Simulate and forecast a random walk
56(1)
Forecast the daily closing price of GOOGL
57(1)
Forecast the daily closing price of a stock of your choice
57(2)
Part 2 Forecasting with statistical models
59(172)
4 Modeling a moving average process
61(20)
4.1 Defining a moving average process
63(6)
Identifying the order of a moving average process
64(5)
4.2 Forecasting a moving average process
69(9)
4.3 Next steps
78(1)
4.4 Exercises
79(2)
Simulate an MA(2) process and make forecasts
79(1)
Simulate an MA(q) process and make forecasts
80(1)
5 Modeling an autoregressive process
81(20)
5.1 Predicting the average weekly foot traffic in a retail store
82(2)
5.2 Defining the autoregressive process
84(1)
5.3 Finding the order of a stationary autoregressive process
85(7)
The partial autocorrelation function (PACF)
89(3)
5.4 Forecasting an autoregressive process
92(6)
5.5 Next steps
98(1)
5.6 Exercises
99(2)
Simulate an AR(2) process and make forecasts
99(1)
Simulate an AR(p) process and make forecasts
100(1)
6 Modeling complex time series
101(39)
6.1 Forecasting bandwidth usage for data centers
102(3)
6.2 Examining the autoregressive moving average process
105(1)
6.3 Identifying a stationary ARMA process
106(5)
6.4 Devising a general modeling procedure
111(14)
Understanding the Akaike information criterion (AIC)
113(1)
Selecting a model using the AIC
114(2)
Understanding residual analysis
116(5)
Performing residual analysis
121(4)
6.5 Applying the general modeling procedure
125(7)
6.6 Forecasting bandwidth usage
132(4)
6.7 Next steps
136(1)
6.8 Exercises
137(3)
Make predictions on the simulated ARMA(1, 1) process
137(1)
Simulate an ARMA(2, 2) process and make forecasts
137(3)
7 Forecasting non-stationary time series
140(16)
7.1 Defining the autoregressive integrated moving average model
142(1)
7.2 Modifying the general modeling procedure to account for non-stationary series
143(2)
7.3 Forecasting a non-stationary times series
145(9)
7.4 Next steps
154(1)
7.5 Exercises
154(2)
Apply the ARIMA(p, d, q) model on the datasets from chapters 4, 5, and 6
154(2)
8 Accounting for seasonality
156(24)
8.1 Examining the SARIMA(p, d, q) (P, D, Q)m model
157(3)
8.2 Identifying seasonal patterns in a time series
160(3)
8.3 Forecasting the number of monthly air passengers
163(15)
Forecasting with an ARIMA(p, d, q) model
165(6)
Forecasting with a SARIMA(p, d, q)(P, D, Q)m model
171(5)
Comparing the performance of each forecasting method
176(2)
8.4 Next steps
178(1)
8.5 Exercises
178(2)
Apply the SAPJMA(p, d, q)(P, D, Q)m model on the Johnson & Johnson dataset
178(2)
9 Adding external variables to our model
180(17)
9.1 Examining the SARIMAX model
182(4)
Exploring the exogenous variables of the US macroeconomics dataset
183(2)
Caveat for using SARIMAX
185(1)
9.2 Forecasting the real GDP using the SARIMAX model
186(9)
9.3 Next steps
195(1)
9.4 Exercises
195(2)
Use all exogenous variables in a SARIMAX model to predict the real GDP
195(2)
10 Forecasting multiple time series
197(19)
10.1 Examining the VAR model
199(2)
10.2 Designing a modeling procedure for the VAR(p) model
201(2)
Exploring the Granger causality test
201(2)
10.3 Forecasting real disposable income and real consumption
203(11)
10.4 Next steps
214(1)
10.5 Exercises
214(2)
Use a VARMA model to predict realdpi and realcons
214(1)
Use a VARMAX model to predict realdpi and realcons
215(1)
11 Capstone: Forecasting the number of antidiabetic drug prescriptions in Australia
216(15)
11.1 Importing the required libraries and loading the data
218(1)
11.2 Visualizing the series and its components
219(1)
11.3 Modeling the data
220(5)
Performing model selection
222(2)
Conducting residual analysis
224(1)
11.4 Forecasting and evaluating the model's performance
225(4)
11.5 Next steps
229(2)
Part 3 Large-scale forecasting with deep learning
231(128)
Introducing deep learning for time series forecasting
233(1)
12.1 When to use deep learning for time series forecasting
234(1)
12.2 Exploring the different types of deep learning models
234(3)
12.3 Getting ready to apply deep learning for forecasting
237(9)
Performing data exploration
237(4)
Feature engineering and data splitting
241(5)
12.4 Next steps
246(1)
12.5 Exercise
246(2)
13 Data windowing and creating baselines for deep learning
248(22)
13.1 Creating windows of data
249(11)
Exploring how deep learning models are trained for time series forecasting
249(4)
Implementing the Data Window class
253(7)
13.2 Applying baseline models
260(8)
Single-step baseline model
260(3)
Multi-step baseline models
263(3)
Multi-output baseline model
266(2)
13.3 Next steps
268(1)
13.4 Exercises
269(1)
14 Baby steps with deep learning
270(17)
14.1 Implementing a linear model
271(5)
Implementing a single-step linear model
272(2)
Implementing a multi-step linear model
274(1)
Implementing a multi-output linear model
275(1)
14.2 Implementing a deep neural network
276(8)
Implementing a deep neural network as a single-step model
278(3)
Implementing a deep neural network as a multi-step model
281(1)
Implementing a deep neural network as a multi-output model
282(2)
14.3 Next steps
284(1)
14.4 Exercises
285(2)
15 Remembering the past with LSTM
287(18)
15.1 Exploring the recurrent neural network (RNN)
288(2)
15.2 Examining the LSTM architecture
290(5)
The forget gate
291(1)
The input gate
292(2)
The output gale
294(1)
15.3 Implementing the LSTM architecture
295(7)
Implementing an LSTM as a single-step model
295(2)
Implementing an LSTM as a multi-step model
297(2)
Implementing an LSTM as a multi-output model
299(3)
15.4 Next steps
302(1)
15.5 Exercises
303(2)
16 Filtering a time series with CNN
305(15)
16.1 Examining the convolutional neural network (CNN)
306(3)
16.2 Implementing a CNN
309(8)
Implementing a CNN as a single-step model
310(4)
Implementing a CNN as a multi-step model
314(1)
Implementing a CNN as a multi-output model
315(2)
16.3 Next steps
317(1)
16.4 Exercises
318(2)
17 Using predictions to make more predictions
320(9)
17.1 Examining the ARLSTM architecture
321(1)
17.2 Building an autoregressive LSTM model
322(5)
17.3 Next steps
327(1)
17.4 Exercises
328(1)
18 Capstone: Forecasting the electric power consumption of a household
329(30)
18.1 Understanding the capstone project
330(3)
Objective of this capstone project
331(2)
18.2 Data wrangling and preprocessing
333(5)
Dealing with missing data
334(1)
Data conversion
335(1)
Data resampling
335(3)
18.3 Feature engineering
338(4)
Removing unnecessary columns
338(1)
Identifying the seasonal period
339(2)
Splitting and scaling the data
341(1)
18.4 Preparing for modeling with deep learning
342(4)
Initial setup
342(1)
Defining the DataWindow class
343(3)
Utility function to train our models
346(1)
18.5 Modeling with deep learning
346(12)
Baseline models
346(3)
Linear model
349(1)
Deep neural network
350(1)
Long short-term memory (LSTM) model
351(1)
Convolutional neural network (CNN)
351(3)
Combining a CNN with an LSTM
354(1)
The autoregressive LSTM model
355(1)
Selecting the best model
356(2)
18.6 Next steps
358(1)
Part 4 Automating forecasting at scale
359(59)
19 Automating time series forecasting with Prophet
361(35)
19.1 Overview of the automated forecasting libraries
362(1)
19.2 Exploring Prophet
363(2)
19.3 Basic forecasting with Prophet
365(5)
19.4 Exploring Prophet's advanced functionality
370(11)
Visualization capabilities
370(4)
Cross-validation and performance metrics
374(5)
Hyperparameter tuning
379(2)
19.5 Implementing a robust forecasting process with Prophet
381(12)
Forecasting project: Predicting the popularity of "chocolate" searches on Google
381(8)
Experiment: Can SARIMA do better?
389(4)
19.6 Next steps
393(1)
19.7 Exercises
394(2)
Forecast the number of air passengers
394(1)
Forecast the volume of antidiabetic drug prescriptions
394(1)
Forecast the popularity of a keyword on Google Trends
394(2)
20 Capstone: Forecasting the monthly average retail price of steak in Canada
396(14)
20.1 Understanding the capstone project
397(1)
Objective of the capstone project
397(1)
20.2 Data preprocessing and visualization
398(2)
20.3 Modeling with Prophet
400(4)
20.4 Optional: Develop a SARI MA model
404(5)
20.5 Next steps
409(1)
21 Going above and beyond
410(3)
21.1 Summarizing what you've learned
411(2)
Statistical methods for forecasting
411(1)
Deep learning methods for forecasting
412(1)
21 Automating the forecasting process
413(5)
21.2 What if forecasting does not work?
413(2)
21.3 Other applications of time series data
415(1)
21.4 Keep practicing
416(2)
Appendix Installation Instructions 418(3)
Index 421
Marco Peixeiro  is a seasoned data science instructor who has worked as a data scientist for one of Canada's largest banks. He is an active contributor to Towards Data Science, an instructor on Udemy, and on YouTube in collaboration with free CodeCamp.