Machine Learning for Time Series Forecasting with Python

CHAPTER 2How to Design an End-to-End Time Series Forecasting Solution on the Cloud

As we discussed in Chapter 1, “Overview of Time Series Forecasting,” time series forecasting is a method for the prediction of events through a sequence of time, by studying the behavior and performance of past phenomena and assuming that future occurrences will hold similar to historical trends and behaviors.

Nowadays time series forecasting is performed in a variety of applications, including weather forecasting, earthquake prediction, astronomy, finance, and control engineering. In many modern and real-world applications, time series forecasting uses computer technologies, including cloud, artificial intelligence, and machine learning, to build and deploy end-to-end forecasting solutions.

To solve real business problems in the industry, it is essential to have a systematic and well-structured template that data scientists can use as a guideline and can apply it to solve real-world business scenarios. The purpose of this second chapter is to provide an end-to-end systematic guide for time series forecasting from a practical perspective by introducing the following concepts:

• Time Series Forecasting Template: A time series forecast template is a set of tasks that leads from defining a business problem through to the outcome of having a time series forecast model deployed and ready to be consumed externally or across your company.

  In the first section, I will introduce some key techniques for ...