BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//AMSE//Event Calendar//FR
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:event-12408@amse-aixmarseille.fr
DTSTAMP:20260430T191550Z
CREATED:20260430T191550Z
LAST-MODIFIED:20260430T191550Z
STATUS:CONFIRMED
SEQUENCE:0
SUMMARY:phd seminar - Linus Nüsing*\, Mathis Preti**
DTSTART:20250916T091500Z
DTEND:20250916T103000Z
DESCRIPTION:*Bayesian Additive Regression Trees (BART) have been shown to b
 e a flexible\, non-parametric regression approach that captures non-linear 
 interactions between covariates and response variables. Building on a multi
 variate extension of the BART framework\, we propose a non-linear vector au
 toregressive model. To induce sparsity in the model\, we incorporate a Diri
 chlet prior over the splitting variables\, which effectively shrinks the pr
 edictor space by selecting only the most relevant (lagged) variables for bu
 ilding the trees. The resulting model is called SUDART. Through a comprehen
 sive Monte Carlo study\, we demonstrate that the estimated generalized impu
 lse responses converge to their underlying true values for linear as well a
 s non-linear data generating processes. This indicates that the SUDART mode
 l correctly recovers both the dynamic structure and the error covariance ma
 trix of the true model\, highlighting its flexibility with respect to the c
 omplexity of the data. The multivariate extension enables us to incorporate
  various (structural) identification methods beyond the order-dependent rec
 ursive Cholesky decomposition to identify macroeconomic shocks\, including 
 for instance identification based on external instruments. We illustrate th
 e usefulness of the model in an empirical application on monetary policy sh
 ocks.**To delay regulation\, the tobacco industry pioneered scientific lobb
 ying to encourage favorable findings\, direct research agendas\, and repres
 s unfavorable publications. I aim to provide the first empirical analysis o
 f how its $355 million in research funding between 1954 and 1998 affected r
 esearch output\, content\, quality\, and co-authorship networks. By linking
  internal tobacco scientific reports with bibliometric data\, I built a nov
 el database of researchers funded by tobacco companies during that time\, i
 ncluding their publications\, career trajectories\, and collaboration links
 .\\n\\nContact: Alexandre Arnout : alexandre.arnout[at]univ-amu.frPhilippin
 e Escudié : philippine.escudie[at]univ-amu.frArmand Rigotti : armand.rig
 otti[at]univ-amu.fr\n\nPlus d'informations: https://amse-aixmarseille.fr/fr
 /evenements/linus-n%C3%BCsing-mathis-preti
LOCATION:MEGA - Salle Carine Nourry\, 424\, Chemin du Viaduc\, 13080 Aix-en
 -Provence
URL;VALUE=URI:https://amse-aixmarseille.fr/fr/evenements/linus-n%C3%BCsing-mathis-preti
CONTACT:Alexandre Arnout : alexandre.arnout[at]univ-amu.frPhilippine Escudi
 é :&nbsp\;philippine.escudie[at]univ-amu.frArmand Rigotti :&nbsp\;armand.r
 igotti[at]univ-amu.fr
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR