You are here

Research topics

CSSEENBA | AMCH | CSSN

 

CSSE Projects

 

High performance computing for nonlinear coupled problems in solid and fluid mechanics

Abstract
Nonlinear coupled problems governed by partial differential equations in solid and fluid mechanics arise in many engineering and biological applications where multiple fields (displacement, damage, thermal, humidity, electric, etc.) are strongly interacting with each other. The present research topic envisages a critical analysis and development of novel numerical strategies for the solution of nonlinearly coupled boundary value problems within the finite element method. Specifically, implicit and explicit numerical schemes, as well as monolithic and staggered solvers, along with suitable high performance computing strategies, will be developed for a wide range of problems selected for their relevance in industrial applications and failure analysis. Prospective applicants are expected to hold a degree in engineering, mathematics, physics, or computer science.
Keywords

nonlinear partial differential equations; coupled problems; high performance computing; solid mechanics; fluid mechanics.
References

  • Reinoso J, Paggi M, Linder C (2017). Phase field modeling of brittle fracture for enhanced assumed strain shells at large deformations: formulation and finite element implementation. COMPUTATIONAL MECHANICS, vol. 59, p. 981-1001, doi: 10.1007/s00466-017-1386-3
  • Lenarda P, Paggi M, Ruiz Baier R (2017). Partitioned coupling of advection–diffusion–reaction systems and Brinkman flows. JOURNAL OF COMPUTATIONAL PHYSICS, vol. 344, p. 281-302, doi: 10.1016/j.jcp.2017.05.011
  • Lenarda P, Gizzi A, Paggi M (2018). A modeling framework for electro-mechanical interaction between excitable deformable cells. EUROPEAN JOURNAL OF MECHANICS. A, SOLIDS, vol. 72, p. 374-392, doi: 10.1016/j.euromechsol.2018.06.001

Reference Faculty: Marco Paggi (MUSAM) and Mirco Tribastone (SYSMA)

Adhesive and cohesive failures in structural adhesives: the interplay between chemistry and mechanics

Abstract

Structural adhesives are used in many industrial applications and are currently designed to guarantee a prescribed load carrying capacity and optimal sealing of the joint. Failures of such joints can be either cohesive or adhesive. In the former case, the crack pattern takes place across the adhesive material, which has its own thickness. In the latter, the interface between the adhesive and the substrate is the weakest link and it leads to premature delamination. In many intermediate situations, both failure modes are concurrently observed. This research topic aims at fully characterizing such failure modes and at understanding how chemical surface treatments can affect the mechanical response of the joint. Both experimental tests in the MUSAM-Lab and numerical research by exploiting the capabilities of the novel phase-field formulation for fracture coupled with the cohesive zone model for delamination will be conducted. Prospective applicants are expected to hold a degree in applied chemistry, materials science, engineering, physics or mathematics.

Keywords

adhesive and cohesive failures; chemical surface treatments; laboratory testing; computational modelling.
References

Reinoso J, Paggi M (2014). A consistent interface element formulation for geometrical and material nonlinearities. COMPUTATIONAL MECHANICS, vol. 54, p. 1569-1581, doi: 10.1007/s00466-014-1077-2

Paggi M, Reinoso J (2017). Revisiting the problem of a crack impinging on an interface: A modeling framework for the interaction between the phase field approach for brittle fracture and the interface cohesive zone model. COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, vol. 321, p. 145-172, doi: 10.1016/j.cma.2017.04.004

Mariggiò G, Reinoso J, Paggi M, Corrado M (2018). Peeling of thick adhesive interfaces: The role of dynamics and geometrical nonlinearity. MECHANICS RESEARCH COMMUNICATIONS, vol. 94, p. 21-27, doi: 10.1016/j.mechrescom.2018.08.018

Reference Faculty: Marco Paggi (MUSAM)

Contact mechanics between rough surfaces: advanced computational modelling and simulation

Abstract

Roughness plays a key role in surface phenomena such as surface physics (heat and electric transfer, hydrophobicity, etc.), surface chemistry (chemical reactions, diffusion, etc.) and tribology (stress transfer, adhesion, lubrication, etc.). Frontier research topics regard the development of finite element-based computational methods allowing for the simulation of contact problems with multiple fields and nonlinear constitutive relations, taking also into account the emergent behaviour induced by microscopic surface roughness. The present research will exploit the new MPJR finite element framework recently published by Paggi and Reinoso, further extending it to rough surfaces in tangential contact and under the action of multiple fields. Joint co-supervision with Prof. Reinoso will be proposed, allowing for the appointment of a double PhD degree at IMT and at the University of Seville, Spain.

Prospective applicants are expected to hold a degree in engineering, physics or mathematics.

Keywords

contact mechanics; roughness; finite element method; coupled problems; nonlinear constitutive relations.

References

Paggi M, Barber JR (2011). Contact conductance of rough surfaces composed of modified RMD patches. INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER, vol. 54, p. 4664-4672, doi:10.1016/j.ijheatmasstransfer.2011.06.011

Vakis AI, Yastrebov VA, Scheibert J, Nicola L, Dini D, Minfray C, Almqvist A, Paggi M, Lee S, Limbert G, Molinari JF, Anciaux G, Aghababaei R, Echeverri Restrepo S, Papangelo A, Cammarata A, Nicolini P, Putignano C, Carbone G, Stupkiewicz S, Lengiewicz J, Costagliola G, Bosia F, Guarino R, Pugno NM, Müser MH, Ciavarella M (2018). Modeling and simulation in tribology across scales: An overview. TRIBOLOGY INTERNATIONAL, vol. 125, p. 169-199, doi: 10.1016/j.triboint.2018.02.005

Paggi M, Reinoso J (2018). A variational approach with embedded roughness for adhesive contact problems, MECHANICS OF ADVANCED MATERIALS AND STRUCTURES, in press, doi:10.1080/15376494.2018.1525454

Reference Faculty: Marco Paggi (MUSAM)

Optimization of additive manufacturing solutions for higher reliability and durability of composites

Abstract

Additive manufacturing solutions are enabling a new era of design optimization, complexity and functionality for composite structures. With the advent of 3D printing technologies, additive manufacturing solutions have rapidly advanced and reached a state of mainstream adoption, particularly for rapid prototyping. Such technologies are only beginning to penetrate and influence the advanced composites industry. The present research project aims at realizing a comprehensive analysis and a critical comparison of the existing additive manufacturing solutions, with special attention to their specific processes. Research will focus on the issues of reliability and durability of composite components realized by such techniques, exploiting computational mechanics tools to simulate each manufacturing process. Optimization strategies will be also explored in order to improve geometries, material combinations, and process parameters towards maximizing the mechanical performance of composites and their durability. Prospective applicants are expected to hold a degree in engineering or mathematics.

Keywords

additive manufacturing solutions; composites; computational mechanics; computational optimization.
Reference Faculty: Marco Paggi (MUSAM) and Alberto Bemporad (DYSCO)

 

ENBA Projects

 

Scale-invariance in Networks, from Renormalization Group to Aggregation. Theory and application to Biochemical reactions.

Abstract

The property of scale-invariance of a physical system allows to determine new and universal properties in the evolution of the system itself. For example, in the evaluation of the scaling properties of the system Hamiltonian it is possible to determine analytically the value of the critical exponents. Graphs are particularly difficult problem to approach since the small-world effect that limits the decimation procedure to few steps determined by the diameter of the instance available. Nonetheless the problem has been approached already in some limit cases, since the possible applications are incredibly important spanning from computer science to statistical physics and to ecology (trophic species). The candidate will work on theoretical models (spanning trees, cayley trees) and applications from biochemistry and social systems.

References

Eigenvector centrality for characterization of protein allosteric pathways

Christian F. A. Negre, PNAS 115 (52) E12201-E12208 (2018)

Cardelli, L., Tribastone, M., Tschaikowski, M., and Vandin, A. (2015). Forward and backward bisimulations for chemical reaction networks. arXiv preprint arXiv:1507.00163.

Cardelli, L., Tribastone, M., Tschaikowski, M., and Vandin, A. (2017). Maximal aggregation of polynomial dynamical systems. Proceedings of the National Academy of Sciences, 114(38):10029–10034.

J. Cardy Scaling and Renormalization in Statistical Physics OUP (1996)

Reference Faculty: Guido Caldarelli, Diego Garlaschelli, Mirco Tribastone

Fake news propagation and cybersecurity: Diffusion in Disordered media

Abstract

The issue of fake news (and the problems arising from them) recently became of great importance for the society. Indeed, while fake news and propaganda always existed in the history of mankind, the computer revolution that created the present connected world made the problem more and more pressing. This happened because it is now possible for everybody to have access to a global audience and to be able to address political/economic/religious and all the other social issues with a simple access to the Internet. Internet and the network of social contacts of Internet users are distributed with a particularly variate topology so that the study of news propagation and modification is particularly troublesome. The project of this thesis should model the propagation of news and their modification on a specific network, namely that of the Facebook users.

References

Hall, G. & Bialek, W. The statistical mechanics of Twitter. arXiv (2018).

Weng, L., Flammini, A., Vespignani, A. and Menczer, F. [2012], ‘Competition among memes in a world with limited attention’,Scientific Reports2, 1–9.11

Reference Faculty: Guido Caldarelli, Rocco De Nicola, and Fabrizio Silvestri (Facebook UK)

Extension of Fitness model to Epidemics

Abstract

Fitness Model of network formation is one simple statistical model that reproduces the property of scale-invariance of the real data. So far it has not been applied to fast varying networks where the properties of the vertices vary in time as in the case of epidemics. The work of the thesis will start from analytical derivations of model extensions to the validation of the computer result with respect to real data of epidemics.

References

Caldarelli, G., Capocci, A., De Los Rios, P. & Muñoz, M. A. Scale-Free Networks from Varying Vertex Intrinsic Fitness. Phys. Rev. Lett. 89, (2002).

Pastor-Satorras, R. & Vespignani, A. Epidemic Spreading in Scale-Free Networks. Phys. Rev. Lett. 86, 3200–3203 (2001).

Tizzoni, M., Sun, K., Benusiglio, D., Karsai, M. & Perra, N. The Scaling of Human Contacts and Epidemic Processes in Metapopulation Networks. Sci. Rep. 5, 15111 (2015).

Reference Faculty: Guido Caldarelli and Daniela Paolotti (ISI Turin)

Study of Country production from Leontief matrices

Abstract

We shall study the network properties of input-output matrices that is of the interrelation of the production and use of different economical sector within one country and across different countries. The candidate will apply the ideas of the network analysis to country exports to this specific case.

References

Leontief, Wassily W. Input-Output Economics. 2nd ed., New York: Oxford University Press, 1986.

The Product Space Conditions the Development of Nations. C. A. Hidalgo, B. Klinger, A.-L. Barabási, R. Hausmann 317 482-487 (2007)

Reference Faculty: Guido Caldarelli, Massimo Riccaboni and Armando Rungi

Study of the Network of Small Enterprises in Tuscany

Abstract

The candidate will collect and analyse data on the network of small enterprises in Tuscany in order to determine the topology of contacts between the different companies and possibly in order to establish the stability of this system with respect to credit risk. The candidate will also work analytically on models of credit to enterprises.

References

Complex Agent Based Models M. Gallegati Springer 2018

The financial accelerator in an evolving credit network. Domenico Delli Gatti, Mauro Gallegati, Bruce Greenwald, Alberto Russo, Joseph E. Stiglitz. Journal of Economic Dynamics & Control 34 (2010) 1627–1650

Reference Faculty: Guido Caldarelli, Nicola Lattanzi and Fabio Saracco

Prosociality, Cognition and Peer Effects

Abstract

This line of research investigates the relationship between the mode of cognition, the structure of social interactions with peers and the likelihood of prosocial behavior. The fundamental drivers of prosocial behavior, i.e., the "social motives", are likely to be fairly stable over time, since they are mainly shaped by social interactions and peer effects. However, which of the social motives actually has the greatest role in shaping behavior in a given situation is likely to depend on the mode of cognition adopted in that situation by the decision-maker. To investigate this we need both theoretical analysis (with models and simulations) and empirical analysis (with experiments and survey data) in order to explore the role of: attention, cognitive effort, empathy, moral reasoning, intrinsic motivations, experienced and expected social sanctioning. The results of this research line should help us better understand the determinants of prosocial behavior and, in particular, to derive implications for policies designed to foster prosociality.

Keywords

Cooperation, Intuition and deliberation, Reciprocity, Social sanctioning, Social network

References

Thöni, Christian, and Simon Gächter. "Peer effects and social preferences in voluntary cooperation: A theoretical and experimental analysis." Journal of Economic Psychology 48 (2015): 72-88.

Belloc, Marianna, Ennio Bilancini, Leonardo Boncinelli, and Simone D’Alessandro. "Intuition and Deliberation in the Stag Hunt Game" (2019), mimeo

Bilancini, Ennio, and Leonardo Boncinelli. "The co-evolution of cooperation and defection under local interaction and endogenous network formation." Journal of Economic Behavior & Organization 70.1-2 (2009): 186-195.

Main researcher: Ennio Bilancini (AXES)

Research units involved: AXES, MOMILAB, NETWORKS, SYSMA

Evolution of Social Behaviors: Norms and Institutions

Abstract

Social norms are patterns of behavior that are self-enforcing within a group: if everyone is expected to conform to the ruling social norm, then everyone actually wants to conform. Social norms are often sustained by different mechanisms: desire to coordinate, fear of being sanctioned, signaling membership in a group, or simply following a leader. The rule of 50-50 sharecropping is an example of a social norm. Institutions are explicit rules (formal or informal) that are enforced by the behaviors and social norms of a society. Social status is a prominent example of an informal institution, while a caste system is an example of a formal one. Stochastic evolutionary game theory can be applied to study the emergence of social norms and institutions. Also, simulations of complex evolutionary dynamics and agent-based modeling are valuable tools in this regard. Further, the instrumental approach to social preferences can shed light on the evolutionary roots of social norms and institutions.

Keywords

Assortativity, Evolutionary dynamics, Evolutionary game theory, Learning, Stochastic stability

References

Bilancini, Ennio, Leonardo Boncinelli, and Jiabin Wu. "The interplay of cultural intolerance and action-assortativity for the emergence of cooperation and homophily." European Economic Review 102 (2018): 1-18.

Bilancini, Ennio, and Leonardo Boncinelli. "Social coordination with locally observable types." Economic Theory 65.4 (2018): 975-1009.

Bilancini, Ennio, and Leonardo Boncinelli. "Instrumental cardinal concerns for social status in two-sided matching with non-transferable utility." European Economic Review 67 (2014): 174-189.

Main researcher: Ennio Bilancini (AXES)

Research units involved: AXES, NETWORKS, SYSMA

Neuroeconomics of Strategic Interaction

Abstract

Game theory extends the model of an individual decision-maker to the case of multiple interacting decision-makers. Solution concepts predict which action profile will result from the actual playing of a game. The most prominent solution concept is Nash equilibrium which typically requires players to use rationality-based inference, common knowledge of beliefs and rationality. Such kind of reasoning can be cognitively extremely demanding and in many cases implausible. In particular, different levels of recursive thinking (i.e., a player’s mental processing that incorporates thinking about others) are likely to require substantial cognitive effort. Moreover, a growing body of experimental evidence reports frequent non-equilibrium play. Investigating which brain circuits are involved in strategic decision-making can help understand the roots of non-equilibrium play. Actually, neuroscientific evidence suggests that different portions of the prefrontal cortex distinguish high versus low levels of recursive thinking, as well as naive versus sophisticated learning, thus encoding the extent of strategic thinking. The aim of this research line is to develop novel models of strategic decision-making that are consistent with such evidence and to test experimentally both novel and existing models, providing new behavioral and neuroscientific data.

Keywords

Brain correlates of strategic reasoning, Bounded cognition, Game theory, Strategic sophistication, Recursive thinking

References

Alós-Ferrer, Carlos. "A Review Essay on Social Neuroscience: Can Research on the Social Brain and Economics Inform Each Other?." Journal of Economic Literature 56.1 (2018): 234-64.

Griessinger, Thibaud, and Giorgio Coricelli. "The neuroeconomics of strategic interaction." Current Opinion in Behavioral Sciences 3 (2015): 73-79.

Bilancini, Ennio, and Leonardo Boncinelli. "Rational attitude change by reference cues when information elaboration requires effort." Journal of Economic Psychology 65 (2018): 90-107.

Main researcher: Ennio Bilancini (AXES)

Research units involved: AXES, MOMILAB, NETWORKS

Measurement of Strategic Ability

Abstract

Strategic interactions have been studied extensively but so far no specific measure of a decision-maker's strategic ability has gained consensus. This research line aims at developing a framework that allows to measure strategic ability. Convincing conceptualizations of (bounded) rationality and cognitive effort as well as of mentalization (i.e., the construction of beliefs about others’ behavior) are crucial ingredients in the development of such framework. A good measure of the strategic ability of an individual should provide information on the likelihood of success that such an individual typically obtains in activities that involve strategic interaction among multiple decision-makers. To this aim original behavioral data have been extensively collected to construct a dataset of actual strategic behavior in a variety of strategic settings, together with socio-demographic and psychological measures. The analysis of such rich dataset should provide a test for the proposed measures of strategic ability as well as source of inspiration for developing novel behavioral measures.

Keywords

Depth of reasoning, Strategic Quotient, Behavioral game theory, Rationality, Theory of Mind

References

Bilancini, Ennio, Leonardo Boncinelli, and Alan Mattiassi. "Assessing Actual Strategic Behavior to Construct a Measure of Strategic Ability." Frontiers in Psychology (2019) forthcoming

Gill, David, and Victoria Prowse. "Cognitive ability, character skills, and learning to play equilibrium: A level-k analysis." Journal of Political Economy 124.6 (2016): 1619-1676.

Alaoui, Larbi, and Antonio Penta. "Endogenous depth of reasoning." Review of Economic Studies 83.4 (2015): 1297-1333.

Main researcher: Ennio Bilancini (AXES)

Research units involved: AXES, MOMILAB

Behavioral Economics of Health and Wellbeing

Abstract

This line of research investigates how we can improve health and well-being of citizens and populations acting on the organization of health care systems, behaviors of health care professionals, habits of individuals and communities. To this aim, methods and tools of behavioral and experimental economics are paired with expertise from life and health sciences, ergonomics and human factors, quality of life studies.

Keywords

Behavioral health economics, Health incentives, Quality and Safety of health care, Subjective well-being, Support to decision-making

References

Rice, Thomas. "The behavioral economics of health and health care." Annual review of public health 34 (2013): 431-447.

Bellandi, Tommaso, Sara Albolino, Riccardo Tartaglia, and Sebastiano Bagnara. "Human factors and ergonomics in patient safety management." In Handbook of Human Factors and Ergonomics in Health Care and Patient Safety, pp. 698-717. CRC Press, 2016.

Bartolini, Stefano, Ennio Bilancini, Luigino Bruni, and Pierluigi Porta, eds. Policies for happiness. Oxford University Press, 2016.

Main researcher: Ennio Bilancini

Research units involved: AXES

Technological change, soft skills and the future of high skilled jobs

Abstract

Advances in information technology and in the globalization of production have been structurally changing labor markets, posing major new challenges to policy makers. Take the case of the “jobless recovery” after the Great Recession, which has been at the center of a lively debate on the possible remedies to structural unemployment. In fact, recent findings suggest a stagnating job market for high skilled professionals since the 2000s driven by a decline in science, technology, engineering and math (STEM) occupations (Beaudry et al., 2016). Eventually, cognitive skills (social intelligence, flexibility, creativity) are instead becoming a necessary albeit not sufficient condition to find good and well-paid occupations (Deming, 2017). Especially in times of a globalization of economic activities and of an automation of production, soft skills may be crucial for their impact on productivity, economic growth and career perspectives. We argue that a truly interdisciplinary approach is needed to better understand the role of soft and cognitive skills (Bakhshi et al., 2017). For example, how can we measure creativity and teamwork? How do cognitive and soft skills relate to STEM and Humanities? Can we complement expert consensus forecasts with the most recent machine learning techniques to predict future job trends across the different demands for skills? Which policies can we derive for R&D, education, productivity and employment?

Keywords

technological change, soft skills, labor markets

Reference faculty: Massimo Riccaboni, Francesco Serti

References

Bakhshi, H., Downing, J., Osborne, M., & Schneider, P. (2017). The future of skills: employment in 2030. London: Nesta, Oxford-Martin, Pearson.

Beaudry, Paul, David A. Green, and Benjamin M. Sand (2016). “The Great Reversal in the Demand for Skill and Cognitive Tasks”, Journal of Labor Economics, 34, 199–247.

Deming, D. (2017). “The growing importance of social skills in the labor market”, The Quarterly Journal of Economics, 132, 1593-1640.

Technology diffusion, productivity of firms, and endogenous economic growth

Abstract

Recent technological developments (e.g. artificial intelligence, 3D printing, blockchain) are already reshaping the organization of the global economy, while cross-country and within-country differences in productivities and incomes persist. In fact, the adoption of Information and Communication Technologies (ICT) in the latest 15 years seems to coincide with a slowdown in aggregate productivity, especially in more advanced economies (Comin and Mestieri, 2018). This is apparently in contradiction with expectations that investment in innovation and productive knowledge shall foster general welfare and economic growth (Romer, 1990). Several explanations have been proposed leading to different conclusions (Haldane, 2017), ranging from a more pessimistic scenario of a “Great Stagnation” to an optimistic quest for errors of measurement in national accounts. Interestingly, the role of the organization of firms and the competitive environment in which they operate, within and across countries, has been so far relatively neglected. How could firms and markets foster or hinder the diffusion of technology? How could a different organization of firms and markets guarantee that technology improvements keep contributing to “perpetually rising standards of living” (Grossman and Helpman, 1994)?

Keywords

endogenous growth theory, firm-level, technology diffusion

Reference faculty: Massimo Riccaboni, Armando Rungi

References

Andrews, D. Criscuolo C. and Gal P. (2016), “The Best versus the Rest: The Global Productivity Slowdown, Divergence across Firms and the Role of Public Policy”, OECD Productivity Working Papers, No. 5

Comin D. and Mestieri M. (2018). “If Technology Has Arrived Everywhere, Why Has Income Diverged?” American Economic Journal: Macroeconomics, 10 (3): 137-78.

Grossman G. M. and Helpman E. (1994). “Endogenous Innovation in the Theory of Growth,” Journal of Economic Perspectives, American Economic Association, vol. 8(1), pages 23-44.

Haldane A. G. (2017). Productivity Puzzles. Speech held by the Chief Economist of the Bank of England at the London School of Economics on 20 March 2017, available online at https://www.bankofengland.co.uk/speech/2017/productivity-puzzles.

Romer, P. M. (1990). "Endogenous Technological Change," Journal of Political Economy, 98:5, S71–102.

Production networks and firms’ outcomes

Abstract

The recent research on firm-to-firm transactions has shown the importance of the structure of a production network for explaining the outcomes of the firms that participate to its formation (Bernard and Moxnes, 2018). For example, more and better suppliers may affect the performance of downstream firms both in domestic and international production networks (Bernard et al., 2016). Yet information frictions may play a role in how links among firms establish and how firms interact, especially when buyers and sellers are distant from each other and when their products are innovative and complex, in which case reciprocal knowledge and trusted partners seems to facilitate the establishment of more productive trade relationships (Chaney, 2016). Eventually, the choices of participating firms collectively determine an input-output macroeconomic equilibrium that could be based on large differences in size and productivity across an economy (Oberfield, 2018). In a context of interdependence between economic agents, there is a large number of unanswered research questions that we can ask ourselves. To what extent and how a firm’s productivity is affected by variations in the performance of its (direct and indirect) suppliers and customers? What are the determinants of the formation of the trade links between firms and, therefore, of inputs adoption? Do firms use existing links to learn about new productivity-enhancing links? Is network closure/clustering between firms important in stimulating cooperation along the supply chain and, therefore, in fostering the adoption and the efficient exploitation of innovative production techniques? What is the role of multinational enterprises in production networks?

Keywords

production networks, network formation, firm-level outcomes

Reference faculty: Massimo Riccaboni, Armando Rungi, Francesco Serti

References

Bernard, A. and Moxnes, A. (2018). Networks and trade. Annual Review of Economics, vol. 10:65-85

Bernard, A., A. Moxnes and Yukiko U. Saito, (2016). Production Networks, Geography and Firm Performance. CEP Discussion Papers dp1435, Centre for Economic Performance, LSE.

Chaney, T. (2016). Networks in international trade. In Y. Bramoulle, A. Galeotti and B. Rogers (eds.), Oxford Handbook of the Economics of Networks, Oxford: Oxford University Press.

Oberfield, E. (2018). A theory of input–output architecture. Econometrica, 86 (2), 559–589.

Machine learning and econometrics

Abstract

A recent trend of research aims to combine methods from machine learning and econometrics for the analysis of economic and business data (Varian, 2014). Typically, machine learning exploits training data and prior knowledge to teach a model how to predict - in the best possible way, given the information available - an output variable as a function of an input vector (Vapnik, 1998). However, classical machine learning methods do not address other issues such as endogeneity and causal inference (Imbens and Rubin, 2015), which have been the traditional field of study of econometrics. In fact, causal inference is particularly important for the analysis of micro/macroeconomic data, and eventually for policy evaluation (Athey and Imbens, 2017). According to (Varian, 2014), machine learning can learn from econometrics how: to deal with data that are not independent and identically distributed; to perform causal inference; to work with instrumental variables. Similarly, econometrics can learn from machine learning how: to train/validate/test and avoid overfitting; to apply non-linear estimation methods such as neural networks; to perform variable selection; to work with extremely large datasets. An example of the successful interaction between the two disciplines is provided by the recent causal tree algorithm (Athey and Imbens, 2016) for the estimation of heterogeneous causal effects. Several other algorithms are expected to be developed in the future.

Keywords

machine learning, econometrics, causal inference

Reference faculty: Massimo Riccaboni, Giorgio Gnecco, Armando Rungi

References

Athey, S., Imbens, G. W. (2016). Recursive partitioning for heterogeneous causal effects. Proceedings of the National Academy of Sciences 113:7353-7360.

Athey, S., Imbens, G. W. (2017). The state of applied econometrics: Causality and policy evaluation. Journal of Economic Perspectives 31, 3-32.

Imbens, G.W., and Rubin, D.B. (2015). Causal Inference for Statistics, Social, and Biomedical Sciences. An Introduction. Cambridge University Press.

Vapnik, V. N. (1998). Statistical Learning Theory. Wiley.

Varian, H.R. (2014). Big data: New tricks for econometrics. Journal of Economic Perspectives 28, 3-28.

Financial shocks, firms’ financial constraints, and production networks

Abstract

Modern economies are organized as webs of specialized producers engaged in exchanges of tangible and intangible inputs. Hence, a network perspective can provide insights on how local shocks can propagate heterogeneously across the entire economy, therefore generating aggregate and granular fluctuations, whose magnitude and volatility may in turn depend on the topology of the underlying production network (Carvalho, 2014; Acemoglu et al., 2016). This is particularly true for the nature of the propagation of financial shocks throughout the real economy. An exogenous financial shock affecting a single firm or a category thereof, e.g. at the beginning of a financial crisis, can reverberate through a supply chain (Escaith et al., 2011; Altomonte et al., 2012) increasing the initial magnitude of the shock and the volatility of the adjustments. More in general, financial constraints at the firm level can transmit heterogeneously across national borders and affect international business cycles (di Giovanni et al., 2018). How can we estimate the direct and indirect impact of a future shock on the real economy? How much responsive are international business cycles to firms’ financial constraints? Which policies could be enhanced for limiting the damage on the general welfare and alleviate the firms’ financial constraints?

Keywords

financial constraints, production networks, global value chains

Reference faculty: Massimo Riccaboni, Armando Rungi

References

Acemoglu D., Carvalho V. M., Ozdaglar A., Tahbaz‐Salehi A. (2016). “The Network Origins of Aggregate Fluctuations”, Econometrica, vol. 80(5): 1977-2016.

Altomonte C., Di Mauro F., Ottaviano G., Rungi A and Vicard V. (2013) “Global value chains during the Great Trade collapse: a bullwhip effect” in: Firms in the international economy: firm heterogeneity meets international business. CESifo Seminar Series. MIT Press, pp. 277-308

Carvalho, V. M. (2014). “From Micro to Macro via Production Networks.” Journal of Economic Perspectives, 28 (4): 23-48.

di Giovanni J, Levchenko A. J., and Mejean I. (2018). "The Micro Origins of International Business-Cycle Comovement," American Economic Review, vol 108(1), pages 82-108.

Escaith H., Lindenberg N. and Miroudot S. (2011). “Global Supply Chains, the Great Trade Collapse and Beyond: More Elasticity or More Volatility?” in “Recovery and Beyond: Lessons for Trade Adjustment and Competitiveness" Editors: F. di Mauro et B. Mandel. European Central Bank.

Interacting Reinforced Stochastic Processes (RSPs) and their applications

Abstract:

Many scientific fields are interested in the stochastic evolution of networks of interacting agents that develop asymptotically a common behavior, a phenomenon typically denoted as “synchronization”. Examples of this type of systems can be found in a large number of scientific areas, such as economics, neuroscience, social and computer science. In all these research areas, the main goals are: (1) to figure out when a synchronization may emerge and (2) to investigate the interplay between the network topology and the collective dynamics followed by the agents. Recently these two goals have been accomplished for networks of agents whose behavior can be modeled by Stochastic Processes that evolve following a Reinforcement mechanism (RSPs), which is when the probability of occurrence of a given event increases with the number of its occurrences in the past. A well-known example of RSP is the Pòlya urn. From a theoretical point of view, this project aims at studying other dynamics for the agents' behaviors and other mechanisms of interaction among them, and, from an applicative point of view, aims at employing the obtained results in order to analyze real interacting systems. This project requires a candidate with strong skills in probability theory and mathematical analysis and statistics.

Keywords

Interacting Random Systems; Reinforced Stochastic Processes; Urn Models; Complex Networks; Preferential Attachment; Weighted Empirical Means; Synchronization; Asymptotic Normality.

Reference Faculty: Irene Crimaldi (AXES)

References

G. Aletti - I. Crimaldi - A. Ghiglietti, Networks of reinforced stochastic processes: asymptotics for the empirical means, forthcoming in Bernoulli.

I. Crimaldi - P. Dai Pra - P-Y. Louis - I. G. Minelli (2019), Synchronization and functional central limit theorems for interacting reinforced random walks, Stochastic Processes and their Applications, 129(1), 70-101.

G. Aletti - I. Crimaldi - A. Ghiglietti (2017), Synchronization of reinforced stochastic processes with a network-based interaction, The Annals of Applied Probability, 27(6), 3787-384