BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:6609@i2m.univ-amu.fr
DTSTART;TZID=Europe/Paris:20201207T140000
DTEND;TZID=Europe/Paris:20201207T160000
DTSTAMP:20241120T201752Z
URL:https://www.i2m.univ-amu.fr/evenements/p-lambert-laplace-p-splines-for
 -approximate-bayesian-inference/
SUMMARY:Philippe Lambert (Université Catholique de Louvain): Oswaldo Gress
 ani : Laplace-P-splines for approximate Bayesian inference
DESCRIPTION:Philippe Lambert: \n\n\nIn Bayesian statistics\, a general and 
 widely used approach to extract information from (complex) posterior distr
 ibutions relies on Markov chain Monte Carlo (MCMC) methods. Although MCMC 
 samplers provide powerful tools for Bayesian inference in various applicat
 ions\, they are often computationally intensive due to their iterative nat
 ure. We propose a much faster alternative for approximate Bayesian inferen
 ce called ‘‘Laplace-P-splines’’ (LPS) that combines Laplace approx
 imations to selected posterior distributions and P-splines for flexible mo
 deling of smooth model terms. After presenting the main ideas underlying t
 he LPS methodology\, we show how it can be used for sampling-free approxim
 ate Bayesian inference in models for survival data. Moreover\, we also mot
 ivate the use of LPS in the class of generalized additive models where the
  response has a distribution belonging to the one-parameter exponential fa
 mily. The proposed methodology is endowed with closed form expressions for
  the gradient and Hessian of the (log) posterior penalty vector. This perm
 its a fast and efficient selection of the penalty parameters tuning the sm
 oothness of the functionals modeled with B-splines. Finally\, the associat
 ed blapsr software package (https://www.blapsr-project.org) is presented t
 o illustrate the use of LPS for inference in latent Gaussian models.\nRefe
 rences\n[1] Gressani\, O. and Lambert\, P. (2018). Fast Bayesian inference
  using Laplace approximations in a flexible promotion time cure model base
 d on P-splines. Computational Statistics and Data Analysis\, 124\, 151-167
 . https://doi.org/10.1016/j.csda.2018.02.007\n[2] Gressani\, O. and Lamber
 t\, P. (2020). The Laplace-P-spline methodology for fast approximate Bayes
 ian inference in additive partial linear models. ISBA Discussion papers. h
 ttp://hdl.handle.net/2078.1/230728\n[3] Gressani\, O. and Lambert\, P. (20
 21). Laplace approximations for fast Bayesian inference in generalized add
 itive models based on P-splines. Computational Statistics and Data Analysi
 s\, 154. https://doi.org/10.1016/j.csda.2020.107088\n\n\n\n
ATTACH;FMTTYPE=image/jpeg:https://www.i2m.univ-amu.fr/wp-content/uploads/2
 020/10/Philippe_Lambert-UCL.jpg
CATEGORIES:Séminaire,Statistique
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20201025T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR