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:20260329T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
TZNAME:EET
DTSTART:20261025T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20260113T121500
DTEND;TZID=Europe/Helsinki:20260113T133000
DTSTAMP:20260711T124010
CREATED:20251222T110035Z
LAST-MODIFIED:20260106T085958Z
UID:18661-1768306500-1768311000@crest.science
SUMMARY:Christophe BRUNEEL-ZUPANC  (KU Leuven) "Don't (fully) exclude me\, it's not necessary! Causal inference with semi-IVs"
DESCRIPTION:[vc_row][vc_column][vc_column_text]Macro seminar\nTime : 12h15 – 13h30 \nDate : 13 th  January 2026 \nSalle 3001 \nChristophe Bruneel-Zupanc (KU Leuven) “Don’t (fully) exclude me\, it’s not necessary! Causal inference with semi-IVs” \nAbstract: This paper proposes semi-instrumental variables (semi-IVs) as a practical alternative to instrumental variables (IVs) to identify the causal effect of a binary (or discrete) endogenous treatment. A semi-IV is a less restrictive form of instrument that is relevant but excluded only from one\, not necessarily both\, potential outcomes. Having two semi-IVs\, one excluded from the treated and the other from the untreated potential outcomes\, is sufficient to nonparametrically point identify the local average treatment effect (LATE) and marginal treatment effect (MTE). In practice\, semi-IVs provide a solution to the problem of finding valid IVs\, because they are easier to find: most selection-specific shocks\, policies\, costs\, or benefits are valid semi-IVs. As an application\, I estimate the returns to manufacturing using sector-specific semi-IVs. \n  \n  \n
URL:https://crest.science/event/christophe-bruneel-zupanc-ku-leuven-t-b-a/
CATEGORIES:Macroeconomics,Seminars
ATTACH;FMTTYPE=:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20260113T140000
DTEND;TZID=Europe/Helsinki:20260113T150000
DTSTAMP:20260711T124010
CREATED:20260108T085000Z
LAST-MODIFIED:20260108T085000Z
UID:18698-1768312800-1768316400@crest.science
SUMMARY:Arnaud GERMAIN (Univ. Catholique de Louvain) "Cluster aggregating: application to Early-Warning System for Non-Performing Clients"
DESCRIPTION:Finance-Insurance\nTime: 14.00 pm\nDate:13th of January 2025\nRoom 3049 \nArnaud GERMAIN (Univ. Catholique de Louvain) “Cluster aggregating: application to Early-Warning System for Non-Performing Clients” \nAbstract : We introduce a new ensemble learning strategy called clagging (for cluster aggregating)\, which consists in combining models fitted on different clusters. First\, we divide the training set into k clusters with k=1\,..\,K. Next\, we fit a model on each of those 1+2+…+K clusters. Finally\, the output for a given test point is obtained by combining the predictions using the distance of the test point to the clusters’ centroids. We perform an extensive horse race study\, considering both regression and classification tasks. Our results suggest that clagging outperforms bagging\, where a bootstrapped sample is traditionally created by drawing observations with replacement until the size of the bootstrapped samples coincides with the size of the original training set. Clagging can also improve the performance compared to a standard fit on the whole training set. \nIn addition\, we apply clagging in the context of default prediction in finance. In its “Guidance to banks on non-performing loans”\, the European Central Bank requires banks to implement an Early Warning System to identify potential non-performing clients at a very early stage. Relying on a unique dataset provided by a systemic European bank\, we show that clagging boost the out-of-sample performance compared to a case where we fit a single prediction model on the whole dataset and a case where we rely on domain knowledge to determine the clusters. \n  \nOrganizers:  Jean-David FERMANIAN \n  \n
URL:https://crest.science/event/arnaud-germain-univ-catholique-de-louvain-cluster-aggregating-application-to-early-warning-system-for-non-performing-clients/
CATEGORIES:Finance-Insurance,Seminars
ATTACH;FMTTYPE=:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20260113T153000
DTEND;TZID=Europe/Helsinki:20260113T164500
DTSTAMP:20260711T124010
CREATED:20260102T150933Z
LAST-MODIFIED:20260106T090457Z
UID:18682-1768318200-1768322700@crest.science
SUMMARY:Vincent ROLLET  (MIT) "Zoning and the Dynamics of Urban Redevelopment"
DESCRIPTION:[vc_row][vc_column][vc_column_text]Macro seminar\nTime : 15h30- 16h45\nDate : 15 th  January 2026 \nSalle 3001 \nVincent ROLLET (MIT) “Zoning and the Dynamics of Urban Redevelopment” \nAbstract: Cities increasingly grow through redevelopment—demolishing old buildings to make way for new ones. This paper studies this process and how it is influenced by zoning\, which regulates the size and uses of new buildings\, using New York City as a case study. I build the first parcel-level panel of a city’s buildings\, zoning\, and floorspace prices. This data allows me to estimate a new dynamic spatial equilibrium model of floorspace supply and demand. I validate the model using quasi-experimental variation from recent zoning reforms and apply it to evaluate the effects of relaxing regulation on construction and affordability. While zoning strongly constrains city growth\, the effects of relaxing regulation take decades to materialize and are limited in inexpensive or densely built areas. This is due to the large fixed costs of redevelopment\, which rise sharply with the size of existing buildings. These costs generate considerable persistence in city structure and substantially lower the expected gains from relaxing zoning. Furthermore\, due to migration\, the affordability benefits of zoning reform largely accrue to households outside the rezoned neighborhoods. \n  \n  \n
URL:https://crest.science/event/vincent-rollet-mit-t-b-a/
CATEGORIES:Macroeconomics,Seminars
ATTACH;FMTTYPE=:
END:VEVENT
END:VCALENDAR