August 2021-July 2026
Grant Agreement ID: 101002867
The proper design of digital platforms as well as of the mechanisms through which platforms compete between them is crucial to avoid waste and enhance social welfare. This research proposal describes three projects that will advance the frontier of our understanding of the working of digital markets. It is motivated by the consideration that the lack of a comprehensive empirical assessment of the crucial phenomena in this area driven by the lack of data availability has been the major impediment to the research in this area. All projects in this proposal touch on the issues of competition digital ad data and privacy and can be orgnazied into three subtopics as follows: Component 1: The Role of Intermediaries in Digital Advertising. This part studies how advertisers and intermediaries select each other by estimating a model of many-to-many matching with transfers. Component 2: Competition Defaults and Antitrust Remedies in Search. This part studies how the antitrust remedies imposed to Google and involving changes in the default settings for search engines on Android devices impacted the penetration of different search engines. Component 3: The Price of Privacy in Digital Markets. This part poses and estimates a structural model of demand and supply for mobile apps to quantify the money for privacy trade-off in consumers' preferences and the corresponding strategic choices by app sellers.
To achieve the project’s goals, a series of activities will be required, the most important of which are:
- Data collection and organization. Various kinds of data will need to be assembled from different sources, including from web scaping via Python.
- The use of both standard econometric techniques and of machine learning methods (both supervised and unsupervised) will be employed to analyze the data (through Python, Stata e R).
- Programming of bidding/pricing algorithms using Q-learners as well as cutting-edge artificial intelligence and machine-learning techniques such as particle filtering and Monte Carlo tree search, one of the building blocks of successful AI systems Deepmind’s AlphaGo.
Articles on Bocconi Knowledge:
This project has been funded by the European Research Council (ERC) under the European Union’s Horizon Europe research and innovation programme.