BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CREST - ECPv5.1.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:CREST
X-ORIGINAL-URL:https://crest.science
X-WR-CALDESC:Events for CREST
BEGIN:VTIMEZONE
TZID:Europe/Helsinki
BEGIN:DAYLIGHT
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:EEST
DTSTART:20250330T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
TZNAME:EET
DTSTART:20251026T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;VALUE=DATE:20250313
DTEND;VALUE=DATE:20250314
DTSTAMP:20260710T082014
CREATED:20250106T140100Z
LAST-MODIFIED:20250106T140100Z
UID:17718-1741824000-1741910399@crest.science
SUMMARY:Julien Trufin (Université Libre de Bruxelles): "Insurance Pricing and Financial Equilibrium through Autocalibration"
DESCRIPTION:Doctoral course: “Insurance Pricing and Financial Equilibrium through Autocalibration” \n10/03/2025 – 13/03/2025 – 17/03/2025 – 20/03/2025 \nReferent: Olivier Lopez \n
URL:https://crest.science/event/julien-trufin-universite-libre-de-bruxelles-insurance-pricing-and-financial-equilibrium-through-autocalibration-2/
CATEGORIES:Actuarial Science,Doctoral Courses
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20250313T100000
DTEND;TZID=Europe/Helsinki:20250313T230000
DTSTAMP:20260710T082014
CREATED:20250304T145351Z
LAST-MODIFIED:20250304T145351Z
UID:17923-1741860000-1741906800@crest.science
SUMMARY:Julien TRUFIN  (Université Libre de Bruxelles) "Bregman and Tweedie Dominances for Candidate Pure Premiums"
DESCRIPTION:[vc_row][vc_column][vc_column_text]Finance & Financial Econometrics : \nTime: 10.00 am\nDate: 13th of March 2025\nRoom 3001 \nJulien TRUFIN (Université Libre de Bruxelles) “Bregman and Tweedie Dominances for Candidate Pure Premiums” \nAbstract :This talk explores Bregman and Tweedie dominances to compare competing candidate pure premiums. An effective testing procedure for Bregman dominance is proposed based on Murphy diagrams and its performance is evaluated through a simulation study. An application to a Swiss motor insurance data set demonstrates the potential of the proposed procedure. A necessary and sufficient condition for Tweedie dominances is then established. Finally\, for autocalibrated predictors\, this talk highlights that Laplace transform order is a sufficient condition for Tweedie dominance. This aligns with the observation that Tweedie dominance is a less stringent concept than Bregman dominance\, which simplifies to the well-known convex order among autocalibrated predictors.\n \nOrganizers: \n\nJean-Michel ZAKOIAN (CREST) \nSponsors:\nCREST[/vc_column_text][/vc_column][/vc_row]\n
URL:https://crest.science/event/julien-trufin-universite-libre-de-bruxelles-bregman-and-tweedie-dominances-for-candidate-pure-premiums/
CATEGORIES:Finance-Insurance,Financial Econometrics,Seminars
ATTACH;FMTTYPE=:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20250313T110000
DTEND;TZID=Europe/Helsinki:20250313T120000
DTSTAMP:20260710T082014
CREATED:20250304T145911Z
LAST-MODIFIED:20250304T151824Z
UID:17924-1741863600-1741867200@crest.science
SUMMARY:Valentina CORRADI (University of Surrey)"Sparsity Tests for High-Dimensional Linear Regression Models in Time Series"
DESCRIPTION:Finance-Insurance\nTime: 11.00 am\nDate: 13th of March 2024\nRoom 3001 \nValentina CORRADI (University of Surrey)”Sparsity Tests for High-Dimensional Linear Regression Models in Time Series” \nAbstract :Penalised Regression methods and in particular the Least Absolute Shrinkage and Selection Operator (LASSO) have become an integral part of modern-day time series analysis. As the performance of LASSO crucially hinges on the assumption of sparsity\, which is unknown in practice\, we propose a Hausman type test to assess it. The null hypothesis of our test is that there are at most k0 relevant regressors with non-zero coefficients. A key distinction between our test and existing methods\, such as Information Criteria\, is that we allow the number of regressors to be (much) larger than the sample size. We further propose a modified version of the test that employs critical values from a moving block bootstrap procedure based on an asymptotic linear representation of the (desparsified) LASSO estimator. We provide two key applications for our test: in the first one\, the test helps to assess the validity of standard Gaussian inference in out-of-sample forecast comparisons with LASSO\, which is inherently related to the presence of sparsity through parameter estimation error. In the second application\, we consider pretesting the validity of Gaussian inference on Impulse Response Functions in high-dimensional (Structural) Vector Autoregressions.\n \nOrganizers:  Zakoian Jean-Michel \n  \n
URL:https://crest.science/event/valentina-corradi-university-of-surreysparsity-tests-for-high-dimensional-linear-regression-models-in-time-series/
CATEGORIES:Finance-Insurance,Seminars
ATTACH;FMTTYPE=:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20250313T121500
DTEND;TZID=Europe/Helsinki:20250313T133000
DTSTAMP:20260710T082014
CREATED:20241218T092419Z
LAST-MODIFIED:20250228T105128Z
UID:17694-1741868100-1741872600@crest.science
SUMMARY:Bernie HOGAN (University of Oxford) - "The Social Implications of Autoencoding"
DESCRIPTION:Sociology Seminar \nTime: :12:15 pm – 13:30 pm \nDate: 13 th of march\nRoom : 3049 \n  \nBernie HOGAN (University of Oxford) – ” The Social Implications of Autoencoding “ \n  \nAbstract :   \nHumans\, like many animals\, are exceptional at detecting and matching patterns in their environment. What makes us distinct\, however\, is our ability to externalize and encode the world through language and technology. For centuries\, technology has been understood as a process of encoding—a deliberate act of representation and storage. Yet\, modern machine learning operates differently : it does not simply encode but autoencodes\, discovering patterns in data without symbolic mediation. \nTo understand the social implications of machine learning\, we must move beyond hallucination discourse and low-level “gotchas” toward a deeper epistemic provocation: autoencoding does not merely extend encoding—it renders encoding epistemically secondary\, unsettling our very assumptions about how knowledge is structured. \nDrawing from research with language models and visual AI\, including work with my colleagues on qualitative studies on participant-aided deepfakes\, I propose a novel framework for the social implications of AI. By foregrounding its abductive reasoning—its ability to infer from the most probable explanation—I argue that autoencoding forces us to rethink not just AI\,  but the epistemic scaffolding that has long defined our relationship to technology. The real social implication is a fundamental shift in what can be made legible—to institutions\, to peers\, and to the systems that govern meaning itself. \n  \nOrganizers:\nPaola TUBARO (Pôle sociologie CREST) \nSofian EL ATIFI (Pôle sociologie CREST) \nPatrick PRÄG (Pôle sociologie CREST) \nSponsors:\nCREST \n
URL:https://crest.science/event/https-www-oii-ox-ac-uk-people-profiles-bernie-hogan/
CATEGORIES:Seminars,Sociology
ATTACH;FMTTYPE=:
END:VEVENT
END:VCALENDAR