Predicting Time-Series Data With Amazon Forecast

Amazon Forecast is a fully managed machine learning service for time-series forecasting. Amazon Forecast uses the same technology as Amazon.com. Accurate forecasting can drive efficiencies and improve business operations. Common examples of time-series data include web traffic and item demand at retail. The data you will use in this lab relates to power consumption data for ten individual households.

 Warning: This lab involves training a model and forecasting which both require around 20 minutes to complete. It is best to perform this lab when you are not constrained by time.

Learning Objectives

Upon completion of this beginner-level lab, you will be able to:

• Understand the main components of Amazon Forecast
• Use Amazon Forecast to forecast structured time-series data
• Explain the trade-offs involved with some key Amazon Forecast settings
• Export Forecast data to CSV files

Intended Audience

• Those interested in gaining insights from their time-series data
• Data Scientists
• Machine Learning Engineers

Prerequisites

Familiarity with the following will be beneficial but is not required:

• Machine learning

The following content can be used to fulfill the prerequisite and obtain a broad understanding of machine learning and its applications:

• Zero to Deep Learning Bootcamp One - Introduction to Data Science and Machine Learning

Updates

April 21st, 2025 - Restored access to Amazon Forecast lab and updated screenshots to reflect latest UI

January 14th, 2022 - Replaced Forecast Lookup (removed by AWS) with a Forecast Export job lab step