Policy Making with Big Data Analytics: Case Studies in Telecoms and Streaming Services

Instructor: Pedro Ferreira, Professor at Carnegie Mellon University jointly in the Heinz College and the Department of Engineering and Public Policy

Course Description: This module introduces participants to decision making using data analytics. The focus is on learning a set of tools that are useful when analyzing a wide range of policy issues.  In this class, these tools will be applied in several telecom-related case studies.  More specifically, participants are introduced to the following issues in telecoms and streaming services:  1) how does changing the length of lock-in periods in triple play services affect firms and consumers? 2) how did the introduction of mobile number portability in Europe affect consumer welfare? 3) how do different recommender systems for streaming platforms affect firms and consumer welfare? Participants are invited to reason about these questions and create plans to use the right data to analytically address them in ways that develop the necessary knowledge to support decision making. These plans include framing and detailing the questions, understanding the data required to answer them, how to measure and collect the data, how to use the data to offer advice for policy making, and how to synthetize, visualize and show results. Case studies using real-world data are presented and discussed.

• Audience: This course is intended for policy analysts who develop reports to support decision making and look for ways to improve how empirical data can be used, collected, and more importantly analyzed to do so. The ideas and the policy tools discussed in this course are general in nature and apply similarly in contexts other than telecom policy.
• Outcomes: Upon completing this course, participants will be equipped with state-of-the-art tools used to develop policy analysis in domains where large datasets are available and where it is important to learn how to correctly anticipate the likely effects of different policy options, i.e., participants will learn how to estimate true causal effects (as opposed to simply identifying correlations in the data).
• Prior Knowledge: Prior knowledge of basic statistics is required. No prior knowledge of data analytics is required.