Python for Time Series Analysis and Forecasting
4.3 (342 ratings)
Course Ratings are calculated from individual students’ ratings and a variety of other signals, like age of rating and reliability, to ensure that they reflect course quality fairly and accurately.
1,974 students enrolled

Python for Time Series Analysis and Forecasting

Work with time series and time related data in Python - Forecasting, Time Series Analysis, Predictive Analytics
4.3 (342 ratings)
Course Ratings are calculated from individual students’ ratings and a variety of other signals, like age of rating and reliability, to ensure that they reflect course quality fairly and accurately.
1,974 students enrolled
Last updated 3/2019
English
English [Auto]
Current price: $72.99 Original price: $104.99 Discount: 30% off
5 hours left at this price!
30-Day Money-Back Guarantee
This course includes
  • 5 hours on-demand video
  • 2 articles
  • 6 downloadable resources
  • Full lifetime access
  • Access on mobile and TV
  • Certificate of Completion
Training 5 or more people?

Get your team access to 4,000+ top Udemy courses anytime, anywhere.

Try Udemy for Business
What you'll learn
  • use Python to perform calculations with time and date based data
  • create models for time series data
  • use models for forecasting
  • identify which models are suitable for a given dataset
  • visualize time series data
  • create ARIMA and exponential smoothing models
  • know how to interpret given models
  • understand time series statistics such as autocorrelation or stationarity
  • use machine learning and deep learning for time series
  • know the alternatives to qualitative methods
  • know how to read a time series plot and understand it (trend, seasonality, constant mean and variance)
Requirements
  • computer with Python and Anaconda ready to use
  • basic Python and Anaconda knowledge (installing packages, Anaconda usage, Python basics)
  • no statistics or time series knowledge is required before the course, you will learn all the relevant things!
  • time and patience to reproduce the examples and to solve the exercises
Description

Use Python to Understand the Now and Predict the Future!

Time series analysis and forecasting is one of the key fields in statistical programming. It allows you to

  • see patterns in time series data

  • model this data

  • finally make forecasts based on those models

  • and of of this you can now do with the help of Python

Due to modern technology the amount of available data grows substantially from day to day. Successful companies know that. They also know that decisions based on data collected in the past, and modeled for the future, can make a huge difference. Proper understanding and training in time series analysis and forecasting will give you the power to understand and create those models. This can make you an invaluable asset for your company/institution and will boost your career!

  • What will you learn in this course and how is it structured?

First of all we will discuss the general idea behind time series analysis and forecasting. It is important to know when to use these tools and what they actually do.

After that you will learn about statistical methods used for time series. You will hear about autocorrelation, stationarity and unit root tests.   You will also learn how to read a time series chart. This is a crucial skill because things like mean, variance, trend or seasonality are a determining factor for model selection.

We will also create our own time series charts including smoothers and trend lines.

Then you will see how different models work, how they are set up in Python and how you can use them for forecasting and predictive analytics. Models taught are: ARIMA, exponential smoothing, seasonal decomposition and simple models acting as benchmarks. Of course all of this is accompanied by homework assignments.

  • Where are those methods applied?

In nearly any field you will see those methods applied. Especially econometrics and finance love time series analysis. For example stock data has a time component which makes this sort of data a prime target for forecasting techniques. But of course also in academia, medicine, business or marketing  techniques taught in this course are applied.

  • Is it hard to understand and learn those methods?

Unfortunately learning material on Time Series Analysis Programming in Python is quite technical and needs tons of prior knowledge to be understood.

With this course it is the goal to make modeling and forecasting as intuitive and simple as possible for you.

While you need some knowledge in maths and Python, the course is meant for people without a major in a quantitative field. Basically anybody dealing with time data on a regular basis can benefit from this course.

  • How do I prepare best to benefit from this course?

It depends on your prior knowledge. But as a rule of thumb you should know how to handle standard tasks in Python.

Who this course is for:
  • data analysts working with time series data (which is essentially any data analyst at some point in the career)
  • people using Python
  • this course is for people working in various fields like (and not limited to): academia, marketing, business, econometrics, finance, medicine, engineering and science
  • generally if you have time series data on your table and you do not know what to do with it and Python, take this course!
Course content
Expand all 46 lectures 05:11:26
+ Introduction
7 lectures 42:21
The Basics of Time Series Analysis and Forecasting
14:06
Select a Forecasting Method
05:56
The Steps of Forecasting - A Guide for Newbies
10:08

Check out if you understood some of the main concepts of this section.

Intro Quiz
5 questions
Python Script to Download
00:07
+ Time Series Analysis Background Knowledge
9 lectures 01:12:13
Time Series Fundamentals
01:55
The Lynx Dataset
02:45
Time Series Vectors and Lags
08:29
Recognizing Time Series Characteristics
15:04
Stationarity
09:54
Autocorrelation
09:45
Visualizing Time Series Data
05:46
Moving Averages and Smoothers
13:10

Check your knowledge on time series statistics and the general background.

TSA Background Quiz
5 questions
Homework Assignment #1: US Inflation Rates
05:25
+ ARIMA for Univariate, Non-Seasonal Data
8 lectures 01:04:53
Introduction to ARIMA Models in Python
02:05
ARIMA Models for Univariate Time Series
10:14
ARIMA Parameter Selection
15:33
ARIMA Residuals
09:20
Manual ARIMA Model Calculation
12:28
Identify ARIMA Model Parameters: General Rules
06:17
ARIMA Forecasts
05:07
Homework Assignment #2: Singapore LFPR
03:49
+ Models for Seasonal Data
11 lectures 01:14:08
The Nottem Dataset
03:27
Seasonal Decomposition
11:23
The STLDecompose Package
00:24
Seasonal Adjustment and Forecasting
07:12
Quantitative Forecasting Methods: An Overview
07:13
Exponential Smoothing Background
11:04
Exponential Smoothing Demo
07:56
What to Do When Numbers Are Not Enough: Qualitative Forecasting Methods
11:50
Introduction to Prophet by Facebook
05:33
Modeling and Forecasting Seasonal Data with Prophet
05:38
Homework Assignment #3: Seasonal Models
02:28
+ Multivariate Time Series Analysis
8 lectures 39:38
Introduction to Multivariate Time Series Analysis and Dataset Structure
10:37
Our Multivariate Time Series Dataset (and Script)
02:48
Checking for Stationarity and Differencing the MTS
03:03
Vector Autoregressive Models
04:45
Fitting a VAR Model and Identifying the Lag Order
05:18
The Granger Causality Test
04:24
Forecasting with VAR Models
04:49
Further References
03:54
+ Homework Solutions
3 lectures 18:12
Homework Solution #1: US Inflation Rates
05:57
Homework Solution #2: Singapore LFPR
07:05
Homework Solution #3: Seasonal Models
05:10