CF

Time Series Analysis, Forecasting, and Machine Learning

22h 47m 45s
English
Paid

Unveiling a new era in time series analysis, this course is designed to transcend traditional methods and immerse you in revolutionary advancements like deep learning and time series classification. From interpreting smartphone data for user insights to decoding brain activity, this course offers an expansive skill set.

Importance of Time Series Analysis

Time Series Analysis has gained tremendous significance in the modern world due to various reasons:

  • The rising inflation has led individuals to explore the stock market and cryptocurrencies to maintain the value of their savings.

  • The COVID-19 pandemic underscored the critical role of forecasting in guiding public health decisions.

  • Businesses are leveraging forecasts to enhance efficiency in inventory management and operational planning.

Covered Techniques

Participants will master a range of techniques, including:

  • ETS and Exponential Smoothing

  • Holt’s Linear Trend Model and Holt-Winters Model

  • ARIMA, SARIMA, SARIMAX, and Auto ARIMA

  • Analysis of ACF and PACF

  • Vector Autoregression and Moving Average Models (VAR, VMA, VARMA)

  • Machine Learning Models: Logistic Regression, Support Vector Machines, Random Forests

  • Deep Learning Models: Artificial Neural Networks, Convolutional Neural Networks, Recurrent Neural Networks

  • Implementation of GRUs and LSTMs for Time Series Forecasting

Practical Applications

The course explores real-world applications such as:

  • Forecasting sales figures using time series data

  • Predicting stock prices and returns

  • Classifying smartphone data to infer user behavior patterns

Exclusive VIP Content

The VIP version offers additional exclusive topics, including:

  • AWS Forecast, Amazon’s advanced low-code forecasting API

  • GARCH, focusing on financial volatility modeling

  • FB Prophet, Facebook’s renowned time series library

About the Author: Udemy

Udemy thumbnail

Udemy is the largest open marketplace for online courses on the internet. Founded in 2010 by Eren Bali, Oktay Caglar, and Gagan Biyani and headquartered in San Francisco, the company went public on the Nasdaq in 2021 under the ticker UDMY. The platform hosts well over two hundred thousand courses across software development, IT and cloud, data science, design, business, marketing, and creative skills, taught by tens of thousands of independent instructors. Roughly seventy million learners use it worldwide, and the corporate arm — Udemy Business — supplies a curated subset of that catalog to enterprise customers.

Because Udemy is a marketplace rather than a single editorial publisher, the catalog is uneven by design. The strongest material lives in the long-form, project-based courses authored by working engineers — full-stack JavaScript, React, Node.js, Python data science, AWS, Docker and Kubernetes, mobile development with Flutter and React Native, and cloud certification preparation. The CourseFlix listing under this source is the slice of that catalog that has been mirrored here for offline-friendly viewing, organized by topic and updated as new releases land. Pricing on Udemy itself swings dramatically with the site's near-permanent sales, which is why the platform is best treated as a deep reference catalog: pick instructors with strong reviews and a track record of updating their material rather than buying on the headline price alone.

Watch Online 170 lessons

This is a demo lesson (10:00 remaining)

You can watch up to 10 minutes for free. Subscribe to unlock all 170 lessons in this course and access 10,000+ hours of premium content across all courses.

View Pricing
0:00
/
#1: Introduction and Outline
All Course Lessons (170)
#Lesson TitleDurationAccess
1
Introduction and Outline Demo
05:26
2
Warmup (Optional)
04:34
3
Where to Get the Code
08:00
4
How to use Github & Extra Coding Tips (Optional)
08:57
5
Time Series Basics Section Introduction
04:19
6
What is a Time Series?
04:53
7
Modeling vs. Predicting
02:31
8
Why Do We Care About Shapes?
05:57
9
Types of Tasks
06:34
10
Power, Log, and Box-Cox Transformations
06:04
11
Power, Log, and Box-Cox Transformations in Code
06:06
12
Forecasting Metrics
11:23
13
Financial Time Series Primer
11:03
14
Price Simulations in Code
03:07
15
Random Walks and the Random Walk Hypothesis
14:36
16
The Naive Forecast and the Importance of Baselines
06:46
17
Naive Forecast and Forecasting Metrics in Code
07:15
18
Time Series Basics Section Summary
03:15
19
Suggestion Box
03:11
20
Exponential Smoothing Section Introduction
03:03
21
Exponential Smoothing Intuition for Beginners
05:38
22
SMA Theory
03:37
23
SMA Code
08:42
24
EWMA Theory
11:08
25
EWMA Code
07:40
26
SES Theory
10:14
27
SES Code
11:56
28
Holt's Linear Trend Model (Theory)
07:56
29
Holt's Linear Trend Model (Code)
03:14
30
Holt-Winters (Theory)
11:21
31
Holt-Winters (Code)
07:53
32
Walk-Forward Validation
09:07
33
Walk-Forward Validation in Code
08:30
34
Application: Sales Data
05:01
35
Application: Stock Predictions
05:38
36
SMA Application: COVID-19 Counting
03:07
37
SMA Application: Algorithmic Trading
02:09
38
Exponential Smoothing Section Summary
04:00
39
(Optional) More About State-Space Models
11:23
40
ARIMA Section Introduction
05:19
41
Autoregressive Models - AR(p)
12:52
42
Moving Average Models - MA(q)
03:32
43
ARIMA
10:46
44
ARIMA in Code
19:16
45
Stationarity
13:02
46
Stationarity in Code
09:51
47
ACF (Autocorrelation Function)
10:11
48
PACF (Partial Autocorrelation Funtion)
06:56
49
ACF and PACF in Code (pt 1)
08:27
50
ACF and PACF in Code (pt 2)
07:04
51
Auto ARIMA and SARIMAX
09:42
52
Model Selection, AIC and BIC
09:51
53
Auto ARIMA in Code
14:05
54
Auto ARIMA in Code (Stocks)
15:46
55
ACF and PACF for Stock Returns
07:02
56
Auto ARIMA in Code (Sales Data)
09:46
57
How to Forecast with ARIMA
09:15
58
Forecasting Out-Of-Sample
01:27
59
ARIMA Section Summary
03:32
60
Vector Autoregression Section Introduction
02:31
61
VAR and VARMA Theory
13:12
62
VARMA Code (pt 1)
07:37
63
VARMA Code (pt 2)
06:48
64
VARMA Code (pt 3)
06:26
65
VARMA Econometrics Code (pt 1)
07:52
66
VARMA Econometrics Code (pt 2)
09:18
67
Granger Causality
04:29
68
Granger Causality Code
03:20
69
Converting Between Models (Optional)
11:46
70
Vector Autoregression Section Summary
03:40
71
Machine Learning Section Introduction
03:53
72
Supervised Machine Learning: Classification and Regression
14:27
73
Autoregressive Machine Learning Models
07:35
74
Machine Learning Algorithms: Linear Regression
05:06
75
Machine Learning Algorithms: Logistic Regression
06:55
76
Machine Learning Algorithms: Support Vector Machines
10:03
77
Machine Learning Algorithms: Random Forest
06:53
78
Extrapolation and Stock Prices
08:48
79
Machine Learning for Time Series Forecasting in Code (pt 1)
13:01
80
Forecasting with Differencing
04:22
81
Machine Learning for Time Series Forecasting in Code (pt 2)
06:48
82
Application: Sales Data
05:25
83
Application: Predicting Stock Prices and Returns
04:53
84
Application: Predicting Stock Movements
04:07
85
Machine Learning Section Summary
02:24
86
Artificial Neural Networks: Section Introduction
03:25
87
The Neuron
09:59
88
Forward Propagation
09:41
89
The Geometrical Picture
09:44
90
Activation Functions
17:19
91
Multiclass Classification
08:42
92
ANN Code Preparation
11:57
93
Feedforward ANN for Time Series Forecasting Code
10:16
94
Feedforward ANN for Stock Return and Price Predictions Code
08:51
95
Human Activity Recognition Dataset
05:54
96
Human Activity Recognition: Code Preparation
06:24
97
Human Activity Recognition: Data Exploration
07:36
98
Human Activity Recognition: Multi-Input ANN
11:00
99
Human Activity Recognition: Feature-Based Model
05:57
100
Human Activity Recognition: Combined Model
03:07
101
How Does a Neural Network "Learn"?
10:50
102
Artificial Neural Networks: Section Summary
02:19
103
CNN Section Introduction
03:08
104
What is Convolution?
16:39
105
What is Convolution? (Pattern-Matching)
05:57
106
What is Convolution? (Weight Sharing)
06:56
107
Convolution on Color Images
15:59
108
Convolution for Time Series and ARIMA
05:00
109
CNN Architecture
23:22
110
CNN Code Preparation
06:17
111
CNN for Time Series Forecasting in Code
06:46
112
CNN for Human Activity Recognition
06:23
113
CNN Section Summary
03:15
114
RNN Section Introduction
04:47
115
Simple RNN / Elman Unit (pt 1)
09:21
116
Simple RNN / Elman Unit (pt 2)
09:43
117
Aside: State Space Models vs. RNNs
03:31
118
RNN Code Preparation
08:39
119
RNNs: Understanding by Implementing (Paying Attention to Shapes)
08:27
120
GRU and LSTM (pt 1)
17:36
121
GRU and LSTM (pt 2)
11:37
122
LSTMs for Time Series Forecasting in Code
09:29
123
LSTMs for Time Series Classification in Code
06:11
124
The Unreasonable Ineffectiveness of Recurrent Neural Networks
03:19
125
RNN Section Summary
02:58
126
GARCH Section Introduction
03:57
127
ARCH Theory (pt 1)
04:58
128
ARCH Theory (pt 2)
07:37
129
ARCH Theory (pt 3)
05:16
130
GARCH Theory
07:41
131
GARCH Code Preparation (pt 1)
07:55
132
GARCH Code Preparation (pt 2)
07:56
133
GARCH Code (pt 1)
06:08
134
GARCH Code (pt 2)
08:31
135
GARCH Code (pt 3)
07:12
136
GARCH Code (pt 4)
05:53
137
GARCH Code (pt 5)
04:21
138
A Deep Learning Approach to GARCH
11:28
139
GARCH Section Summary
06:37
140
AWS Forecast Section Introduction
08:03
141
Data Model
09:17
142
Creating an IAM Role
04:10
143
Code pt 1 (Getting and Transforming the Data)
10:00
144
Code pt 2 (Uploading the data to S3)
12:53
145
Code pt 3 (Building your Model)
06:53
146
Code pt 4 (Generating and Evaluating the Forecast)
06:50
147
AWS Forecast Exercise
02:55
148
AWS Forecast Section Summary
04:56
149
Prophet Section Introduction
03:12
150
How does Prophet work?
08:25
151
Prophet: Code Preparation
12:42
152
Prophet in Code: Data Preparation
09:00
153
Prophet in Code: Fit, Forecast, Plot
08:31
154
Prophet in Code: Holidays and Exogenous Regressors
10:20
155
Prophet in Code: Cross-Validation
06:08
156
Prophet in Code: Changepoint Detection
04:15
157
Prophet: Multiplicative Seasonality, Outliers, Non-Daily Data
10:17
158
(The Dangers of) Prophet for Stock Price Prediction
13:11
159
Prophet Section Summary
03:28
160
Anaconda Environment Setup
20:21
161
How to install Numpy, Scipy, Matplotlib, Pandas, IPython, Theano, and TensorFlow
17:23
162
How to Code by Yourself (part 1)
15:57
163
How to Code by Yourself (part 2)
09:24
164
Proof that using Jupyter Notebook is the same as not using it
12:30
165
How to Succeed in this Course (Long Version)
10:25
166
Is this for Beginners or Experts? Academic or Practical? Fast or slow-paced?
22:05
167
Machine Learning and AI Prerequisite Roadmap (pt 1)
11:19
168
Machine Learning and AI Prerequisite Roadmap (pt 2)
16:08
169
What is the Appendix?
02:49
170
BONUS Lecture
05:32
Unlock unlimited learning

Get instant access to all 169 lessons in this course, plus thousands of other premium courses. One subscription, unlimited knowledge.

Learn more about subscription

Related courses

Frequently asked questions

What are the prerequisites for this course?
The course assumes a foundational understanding of statistics and basic programming skills. Familiarity with Python and data manipulation libraries is beneficial as the course involves coding exercises. Early lessons such as 'How to use Github & Extra Coding Tips' provide additional support for those less experienced in these areas.
What projects or practical applications will I work on?
Students will engage in practical applications including sales data forecasting, stock prediction, COVID-19 case counting using Simple Moving Averages, and algorithmic trading. These projects provide hands-on experience with real-world data, enhancing understanding of the techniques covered.
Who is the target audience for this course?
This course is ideal for data scientists, analysts, and professionals in finance or public health who seek to enhance their forecasting skills using time series analysis. It is also suitable for those interested in applying machine learning and deep learning to interpret time-dependent data across various domains.
How does the depth of this course compare to other courses on time series analysis?
The course covers both traditional and advanced techniques in time series analysis, including deep learning models like Recurrent Neural Networks and GRUs. It offers broader coverage than many introductory courses by integrating machine learning approaches and real-world applications.
What specific tools or platforms are used in this course?
The course uses Python extensively, with coding exercises involving libraries for time series analysis and machine learning. Lessons such as 'Where to Get the Code' and 'Auto ARIMA in Code' emphasize practical implementation of the techniques discussed.
What topics are not covered in this course?
The course does not cover non-time series specific machine learning models or tools unrelated to forecasting. It focuses exclusively on time series analysis, excluding broader data science techniques not applicable to time-dependent data.
What is the expected time commitment for this course?
With 170 lessons, the course is comprehensive and may require significant time investment, depending on the student's pace. It includes optional sections, allowing learners to tailor their experience. Engaging with coding exercises and projects will also influence the overall time commitment.