I specialize in econometrically rigorous causal machine learning for observational panel data, using Neyman-orthogonal, cross-fitted estimators to identify heterogeneous treatment effects and elasticities under high-dimensional confounding
Methods: Double/Debiased ML (R/X/U/IV-learners), causal forests (grf), debiased lasso, IV-GMM for continuous treatments, panel-DML with FE, modern DiD/event-study corrections (Sun–Abraham; Callaway–Sant’Anna), synthetic controls/matrix-completion, and valid inference via sample-splitting/cross-fitting.
My field focus is global pharmaceutical markets—Ramsey/Boiteux pricing and reimbursement, insurance/copay design, generic entry/substitution, patent expiry, reference pricing/tenders/parallel trade, and the effects of safety recalls and shortages on competition and innovation
I estimate demand (BLP/random-coefficients with valid instruments), recover marginal costs, solve multi-product static/dynamic pricing (MPEC/NFP) and Markov-perfect oligopoly, and run counterfactuals (mergers, policy/pricing rules, recalls) with full welfare decompositions.
-
Designing panel data that identifies: FEGLS-based rules for the budgeted trade-off between more units and cleaner labels, improving DML identification in panels.
-
Predicting city competitiveness: Cluster-then-learn pipeline that raises out-of-sample accuracy and interpretability—transferable to product segmentation and IV construction.
-
Modeling spillover dynamics: A network-reinforcement model that maps innovation spillovers—an IO analogue to cross-price interactions and policy propagation.
Secondary fields
Related ML/innovation projects (cluster-then-learn; network spillovers) that I use to segment markets, build better instruments/proxies, and model cross-product interactions—methods I now apply to identify heterogeneous (cross-)elasticities and run welfare-relevant counterfactuals in pharma IO.
Learning by doing
I program in Python, Matlab, R and STATA with a slight preference for STATA and Python.
Programming skills
Main Field Works
I strongly believe that economic research should be policy relevant. In order to understand what is policy relevant or not, literature reviews alone may be not sufficient. This is why I am working hard to link with industries as well.
Links with industry
I prioritize design and estimation that survive robustness checks—Neyman-orthogonal DML/IV, modern DiD, and placebo/heterogeneity tests—so estimates have a clear causal interpretation.
I pair demand (BLP/random coefficients) with multi-product pricing to recover costs and run policy counterfactuals (Ramsey rules, tenders, reference pricing, recalls) with transparent welfare decompositions.
Cross-fitted DML, debiased regularization, and validated prediction pipelines scale to large product-level panels and deliver heterogeneous (cross-)price elasticities with valid inference.
Whereby possible codes are provided for full reproducibility of the papers.
My main teaching experiences are in Causal Machine-Learning and Introductory IO.
My work is academic first, but I design projects to be policy-relevant and operational. I pressure-test results with real-world constraints in pharma and payer settings—helped by prior experience at Roche Pharma and ongoing engagement with EU-level discussions on advanced therapy medicinal products (ATMPs)—so the questions I ask in Empirical IO and causal ML map to the problems practitioners face (pricing and reimbursement of high-cost therapies, access under budget pressure, evidence standards, tenders/reference pricing). That feedback loop shapes identification choices, instruments, and counterfactuals, and it turns findings into implementable insights.
What people say about Me !
Massimo Riccaboni
Professor of Economics, IMT School for Advanced Studies Lucca
"Federico is one of the best PhD students in our Economics, Networks, and Business Analytics program. His application of Matrix Completion to assess the complexity of countries, particularly in economic contexts, is both novel and promising. I believe his findings will appeal to academics and policymakers, offering valuable insights into economic complexity."
Giorgio Gnecco
Associate Professor in Operations Research, IMT School for Advanced Studies Lucca
"Federico has impressed me with his scientific curiosity and genuine interest in research. His approach has allowed him to be highly productive in a short period across different topics. I am convinced that he is fully qualified for the position in Applied Economics at Bocconi University."
Step into the reflections of those who have shared this intellectual journey with me, Their words paint a picture of our shared exploration of economic complexities, innovation, and industrial dynamics.
Embark on a collaborative journey in academic research. My expertise aligns with your quest for impactful collaborations and research endeavors.
In this dynamic space, I invite you to explore the fusion of economic theories, machine learning innovations, and the future landscape of analytics. Let's navigate the ever-evolving terrain of economics together.
Academic Collaborative Research Endeavors
