BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//AMSE//Event Calendar//FR
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:event-7944@www.amse-aixmarseille.fr
DTSTAMP:20260430T232816Z
CREATED:20260430T232816Z
LAST-MODIFIED:20260430T232816Z
STATUS:CONFIRMED
SEQUENCE:0
SUMMARY:Empirical &amp\; Econometric Methods Session - Mathias Silva Vazque
 z
DTSTART:20210217T150000Z
DTEND:20210217T150000Z
DESCRIPTION:Bayesian parameter inference consists in updating prior beliefs
  about the possible values of a model's parameters through the probability 
 of reproducing the observed data with the model evaluated at each of these 
 values (i.e.\, the likelihood function). For many models\, an expression fo
 r the likelihood function is unknown and therefore many Bayesian methods ar
 e infeasible. Approximate Bayesian Computation (ABC) methods overcome this 
 restriction by exploiting simulations from the model to approximate the unk
 nown likelihood function. For this purpose\, by expliciting what we conside
 r a 'close enough' reproduction of the data in the simulations ABC can appr
 oximate the updated or posterior probability distribution of the model's pa
 rameters. This presentation will cover basic elements of Bayesian inference
 \, an overview of common Markov Chain Monte Carlo (MCMC) sampling methods u
 sed for this purpose\, and an introduction to ABC methods exploiting these 
 notions with applications to common Econometric models.\\n\\nContact: Julie
 ta Peveri: julieta.peveri[at]univ-amu.frBertille Picard: bertille.picard[a
 t]univ-amu.frMathias Silva: mathias.silva-vazquez[at]univ-amu.fr\n\nPlus d'
 informations: https://www.amse-aixmarseille.fr/en/events/mathias-silva-vazq
 uez-2
URL;VALUE=URI:https://www.amse-aixmarseille.fr/en/events/mathias-silva-vazquez-2
CONTACT:Julieta Peveri:&nbsp\;julieta.peveri[at]univ-amu.frBertille Picard:
  bertille.picard[at]univ-amu.frMathias Silva: mathias.silva-vazquez[at]univ
 -amu.fr
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR