A monthly online seminar on the methodology of data science, its applications, and related governance issues
Wednesday February 4 2026 (11AM ET - 5PM Rome time): Tim Lenton, University of Exeter
Title: Positive tipping points to avoid climate tipping points
Abstract: Tipping points in climate science normally refer to small changes in the Earth system that unleash much broader, typically damaging impacts that accelerate climate change. Well-known examples are rising sea levels due to disintegration of the Greenland and West Antarctica ice sheets, or the release of methane from the thawing permafrost. They help to underline the urgency of climate action. Today most people understand we must reduce emissions – and very quickly. In this webinar, Tim will summarise recent evidence regarding climate tipping points, which supports declarations that we are in a ‘climate emergency’. Then he will turn to identifying positive social tipping points that will need to be triggered to have any hope of limiting global warming to well below 2C.
Wednesday March 4 2026 (11AM ET - 5PM Rome time): Sam Baugh, Penn State University
Wednesday April 8, 2026 (11AM ET - 5PM Rome time): Samuele Centorrino, International Monetary Fund
Wednesday May 6, 2026 (11AM ET - 5PM Rome time): Federico Crippa, Northwestern University
Wednesday June 10, 2026 (11AM ET - 5PM Rome time): Tallys Feldens, Sant'Anna School of Advanced Studies
Title: More Care, Less Cash? Unpacking the EU’s caregivers labor market losses in late career
Abstract: Informal caregivers are an important complement to formal healthcare, especially by helping the patients with their daily activities, which prevents institutionalization. However, by providing informal care, people might be giving up paid working hours, which can also affect their propensity to retire when they are in late career. Thus, it is important to ascertain the amount of change in working hours, personal income and transition to retirement that is linked to informal caretaking. Data from 2006 to 2022 of the Survey of Health, Ageing and Retirement in Europe (SHARE) is used to collect the individuals over 50 years who reported to have helped people inside or outside their household with personal care, practical household help or help with paperwork. We explored three datasets: the ones who became a caregiver, the ones who left the caregiver role, and those who left the caregiver role due to the death of the care receiver. We compared these three samples against the ones who were never caregivers and employed an event study estimation. To account for the (possibly) non-random allocation in the choice and ability to become a caregiver, the first step in the analysis entailed a propensity score matching (PSM) selecting only those who retain similar observable characteristics. Becoming a caregiver is associated with lower employment income, raising pensions income and a higher chance of retirement. This trend is not reversed when people leave the caregiver roles: either for the ones who stop caregiving for any reason or for the ones who experienced the death of the cared person, the aggregated dynamic effects point to a reduction in employment income and raise in the pension’s income. We investigated age, gender, place of caregiving, country of residence and inter-household effects. These results indicate that caregiving roles have short to medium-term effects and probably affect the late-career choices of the ones who have someone to take care of. Although the provision of informal care can contribute to social support and even generate savings to the health care system, our estimates show that it has a non-negligible impact on late-career choices. Specific policies should address this important trade-off and estimate incentives to decrease the burden of informal care and its financial implications without compromising productivity and human capital losses.
The seminars are held on Microsoft Teams (link provided through the mailing list) and last 60 minutes:
45 minutes of presentation
15 minutes of Q&A
If there is anyone you would like to hear at the L'EMbeDS Data Science Seminar Series, you may let us know here.
The seminars will span a broad range of topics concerning data science methodology, its applications in various domains, and related governance issues. These include, but are not limited to:
Methodological innovation: Developments in areas such as machine/statistical learning; approaches to enhance reproducibility, interpretability and privacy; causal inference; Bayesian statistics; techniques for high-dimensional and structured data; time series and longitudinal data analysis; spatial statistics; functional data analysis; simulation models and methods; synthetic data generation; etc.
Data- and computation-driven research: Instances of how data science methodology empowers research and facilitate the solution of real-world problems in diverse domains such as economics, finance, development and the study of inequalities; climate change and its environmental and socio-economic impacts; public policy and the study of social systems; STI (Science, Technology & Innovation) studies; biomedical research, public health, and the study of healthcare systems; ageing, demographic shifts and their impacts; etc.
Governance and regulatory frameworks: Examination of how data science intersects with governance challenges and the development of appropriate regulatory responses in areas such as privacy and accountability in data and algorithmic regulation; individual and intellectual property rights; regulation of data-driven technologies, digital markets and platforms; digitalization of public administration and services; cybersecurity; etc.
(seminars providing original literature reviews are also welcome!)
Irina Carnat, Sant'Anna School of Advanced Studies
Matteo Coronese, Sant'Anna School of Advanced Studies
Lorenzo Testa, Carnegie Mellon University
Francesca Chiaromonte, Sant'Anna School of Advanced Studies and Penn State University
We gratefully acknowledge the community of the L'EMbeDS Department of Excellence at Sant'Anna School of Advanced Studies, Pisa, Italy. Our seminar series is inspired by similar initiatives, such as the Online Causal Inference Seminar, the International Seminar in Selective Inference, etc.