Archivio istituzionale della Ricerca - Bocconi
Not a member yet
30225 research outputs found
Sort by
The limits of identification in discrete choice
Get PDFThis paper uncovers tight bounds on the number of preferences permissible in identified random utility models. We show that as the number of alternatives in a discrete choice model becomes large, the fraction of preferences admissible in an identified model rapidly tends to zero. We propose a novel sufficient condition ensuring identification, which is strictly weaker than some of those existing in the literature. While this sufficient condition reaches our upper bound, an example demonstrates that this condition is not necessary for identification. Using our new condition, we show that the classic “Latin Square” example from social choice theory is identified from stochastic choice data
Technological change and domestic outsourcing
No full textDoes domestic outsourcing react to technological change? We study the staggered diffusion of broadband internet in France in the 2000s, and show that connected firms increased their outsourcing expenditures while decreasing the diversity of occupations they employ in-house. Meanwhile, employment in non-core occupations became increasingly concentrated in firms specializing in subcontracting services. Finally, we provide evidence that workers in high-skill occupations experienced salary gains from being outsourced, while workers in low-skill occupations lost out. Overall, we show that the deployment of new technologies stimulated domestic outsourcing in this context, with important implications for labor market inequality
A Longitudinal Study of the Gender Gap in School Grades via Flexible Bayesian Beta Regression
No full textShared decision-making
No full textPatient-centered care aims to empower patients to become active participants in their care, with a shift from provider-centric norms to care processes arranged around patients’ beliefs and needs. Patient centricity calls in fact for accurate consideration and reflection of patients’ values, preferences, and choices at every step in the healthcare pathway. This requires that physicians develop good communication skills and address patient needs effectively [Reynolds, 2009]. Communication is a crucial factor impacting the perceived quality of care from the perspective of patients and their caregivers. An adequate communication is expected to have a positive influence on the quality of care. Charles, Gafni, and Whelan [2004] distinguished four possible archetypes in patient-clinician interactions, which in turn reflect alternative approaches in communication summarized in Table 24.1
Economic evaluations of health service interventions targeting patients with multimorbidities: a scoping literature review
Get PDFIntroduction: Multimorbid patients have been growing, leading to an exponential increase in healthcare costs and patterns of resource utilization. Despite the heightened interest toward integrated care programs as a response to the complex need of multimorbid patients, economic evaluations of these programs remain scarce. This work investigated the economic evaluations of service interventions targeting multimorbid patients, to identify the characteristics of these programs and the methods applied to their evaluation.
Methods: We conducted a scoping review of papers published between 2010 and 2021 on PubMed, Science Direct, EconLit and Web Of Science. The search strategy was built around three keyword blocks: service interventions, multimorbidity, economic evaluations. We selected economic evaluations of service interventions delivered through multiple care settings and targeting patients with 2+ chronic conditions.
Results: Twenty-five articles were included. Interventions were categorized as organizational-type versus patient-oriented. The selected studies often targeted patients with one chronic disease, associated with a mental disorder, like depression or anxiety. Included studies were mostly cost-utility analyses conducted with the healthcare perspective.
Discussions and conclusions: This work confirmed that economic evaluations of service interventions for multimorbid patients are limited in number. This could suggest that decision-making regarding the delivery of healthcare services for multimorbid patients may not always be based on a solid evidence base. More economic analyses are needed to inform evidence-based coverage decision-making
Topics in scalable Bayesian posterior estimation
Get PDFAs the complexity and dimensionality of data continue to increase, it is becoming fundamental to develop advanced strategies for statistical inference and to explore their computational properties (Bishop, 2006).
This thesis considers Bayesian models, known for their ability to frame prediction and uncertainty within a coherent probabilistic framework. However, achieving accurate estimates of posterior quantities within these models generally requires innovative techniques to accommodate the challenges of modern data analysis. We aim at developing algorithms for exact and approximate posterior estimation exhibiting linear computational cost in the number of parameters, for asymptotic settings where both the numbers of parameters and observations grow to infinity. Such performances are substantially unattainable for state of the art gradient based sampling methods, and are achieved only leveraging the hidden probabilistic structure of the models under consideration.
The first and second chapters of this document focus on couplings, a relatively simple probabilistic construction whose potential for unbiased estimation has been recently spotlighted thanks to the work of (Glynn and Rhee, 2014; Jacob et al., 2020). After a brief review on couplings and their applications for unbiased sampling and estimation in Chapter 1, we present in Chapter 2 theoretical results bounding the computational effort required by the coupling construction of Jacob et al. (2020) for certain Gibbs samplers, proving its scalability in a wide range of applications, spanning from crossed random effect to sparse graphical models. Unbiased estimation via couplings therefore presents a promising way to enhance the precision and accuracy of statistical inference, offering insights beyond traditional estimation approaches.
Turning to Chapter 3, we cover topics related to variational inference. Variational inference has captured significant attention in the past decades: essentially, it translates the probabilistic problem of finding the posterior distribution as an optimization task (Blei et al., 2017). This chapter not only presents its theoretical foundations but also explores practical implementation and provides results on scalability of the mean field variational approximation for certain large scale hierarchical models. More in detail, assuming the data is randomly generated from a specific distribution, we characterize the rate at which the iterates produced by the coordinate ascent variational inference (CAVI) algorithm converge to a variational minimizer for large scale hierarchical models, proving dimension-free convergence under warm start assumptions. Our work builds upon (Ascolani and Zanella, 2024), highlighting the effectiveness of CAVI in efficiently approximating posterior quantities for models where Gibbs sampling has proved to be effective, given the inherent similarities between these coordinate-wise schemes (Tan and Nott, 2014).
Chapter 4 contains some recent advances developed during my visiting period at Warwick University with professor Gareth Roberts. Specifically, we study some properties of Catalytic couplings (Breyer and Roberts, 2001), a coupling procedure well suited for settings where only unnormalized distributions are available and able to couple multiple chains at once.
In summary, this dissertation aims at presenting efficient methods in the realm of Bayesian posterior estimation for models with sparse dependencies, such as hierarchical and crossed models, leveraging their probabilistic structure to obtain linear cost estimates. By investigating coupling methods and variational inference, we aim at helping bridge the gap between state-of-the-art statistical procedures and the understanding of their computational properties
Essays in Barriers and Pathways to Equitable Development
Get PDFThis dissertation examines how policy interventions can shape individual behavior and development across three distinct contexts: admission tests, mandatory markets for pension savings, and early childhood development. The first paper explores how textual context in Brazil’s ENEM affects performance gaps by socioeconomic status, gender, and ethnicity, revealing that semantic features in test questions significantly influence disparities and identifying pathways for fairer test design. The second paper evaluates a pro-competition reform in Chile’s pension system that made cheaper options available but saw limited uptake due to high levels of inertia. By exploring the timing and consequences of changes in fees, the study finds that participants are much more sensitive to price increases than to equivalent cost reductions. The third paper, co-authored with Diana Krüger, Matias Berthelon, and Rafael Sánchez, leverages an exogenous shock to breastfeeding caused by an earthquake in Chile to estimate its causal effects on child development, showing significant cognitive benefits of extended breastfeeding. Together, these essays highlight the nuanced ways in which policies can mitigate barriers to equity and promote development, offering evidence-based insights for designing effective interventions
The AI-driven public actor: new challenges to fundamental rights and the role of the private sector
Get PDFIn the last few years, with the increasing use of technology, and especially of Artificial Intelligence (AI), new aspects in the exercise of the public actor’s power can be observed. Specifically, as AI technologies begin to play a dominant role in the contemporary exercise of power, it becomes increasingly important to examine the phenomenology of a new kind of power and its unique challenges to constitutional principles.
On this merit, the public actor is making extensive use of technological and automated tools to reach fast and more efficient decisions. Considering that it has the privilege to acquire, ex lege, both ex officio or on the initiative of citizens, a wide availability of personal data and information, the development of AI technologies for public decision-making processes is now inevitable.
Therefore, the rise of a new paradigm can be observed, the so called ‘AI-driven public actor’, which has elements of absolute novelty.
In this context, this work has the aim of answering two major questions. First, which is the impact of the new paradigm of the ‘AI-driven public actor’ on the exercise of the citizens’ fundamental rights, considering the guarantees that the European Union (EU) legal order is currently trying to ensure. Second, the thesis investigates whether the EU legal order provides for specific provisions to balance the role of the private actor in public decisions.
Accordingly, the answers to these questions have important implications not only in terms of understanding the future of the application of fundamental rights, but also of the maintenance of democratic institutions, as it may depends on the ways in which new technologies are changing the relationship between the public actor, the private one and the civil society
The circular economy and SMEs
No full textThe professional role of circularity in industry performance efficiency encompasses a
multifaceted approach to resource management, innovation, stakeholder engagement, and
operational excellence, all of which contribute to sustainable business practices and enhanced
competitiveness. In this vein, the main aim of this book is to interpret circularity and the circular
economy through different levels of target companies and industries to disclose the
implementable policies and implications. In this regard, the current book tries to investigate
the improvement of the circularity concept as an important tool for a sustainable future.
According to the aforementioned aim, the different chapters of the present book will focus on
the concept of circularity as a solution in ecosystem management, a circular fashion perspective
and measuring circularity metrics for target industries, implementation of the digital ecosystem
to improve the circularity, integration of smart production with the circular economy,
interactions between consumer skepticism and the circular economy, and managerial
propositions and legal solutions for improving the adoption of circular business models in
example industries
Information and Permitting: Merger Review, Drug Approval, and Food Labeling
Get PDFThis dissertation investigates how welfare is shaped by asymmetric information between firms, consumers, and regulators, and how targeted policy interventions can mitigate the resulting distortions across three domains: merger control, drug approval, and food labeling. The first chapter analyzes merger control in a setting where firms have an informational advantage over regulators concerning synergies. While a stronger informational advantage increases total information, the resulting rise in asymmetry reduces welfare—unless the regulator can commit to a decision rule. The second chapter studies drug approval in a setting where firms privately observe product safety and self-select into investment. Welfare improves when the regulator uses price to screen firm entry, whereas relying solely on scientific evaluation criteria proves less effective. The third chapter analyzes food labeling in a setting where some consumers are unaware of key product attributes, showing that mandatory warning labels are most effective at safeguarding consumer welfare because they withhold positive information from the unaware. Together, these essays offer theoretical frameworks for understanding and addressing welfare losses resulting from informational disadvantages on the part of regulators or consumers