BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Local limit theorem for Robbins Monro algorithms
DTSTART;VALUE=DATE-TIME:20181113T083000Z
DTEND;VALUE=DATE-TIME:20181113T093000Z
DTSTAMP;VALUE=DATE-TIME:20220128T033128Z
UID:indico-contribution-3352@indico.math.cnrs.fr
DESCRIPTION:Speakers: Lorick Huang (INSA / IMT)\nThe Robbins-Monro algorit
hm is a recursive\, simulation-based stochastic procedure to approximate t
he zeros of a function that can be written as an expectation. It is known
that under some technical assumptions\, a Gaussian convergence can be esta
blished for the procedure. Here\, we are interested in the local limit the
orem\, that is\, quantifying this convergence on the density of the involv
ed objects. The analysis relies on a parametric technique for Markov chain
s converging to diffusions\, where the drift is unbounded.\n\nhttps://indi
co.math.cnrs.fr/event/3779/contributions/3352/
LOCATION:Toulouse School of Economics Bât S\, amphi MS001
URL:https://indico.math.cnrs.fr/event/3779/contributions/3352/
END:VEVENT
END:VCALENDAR