BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Inference with approximate likelihoods - Helen Ogden (University o
 f Southampton)
DTSTART:20170703T143000Z
DTEND:20170703T151500Z
UID:TALK73131@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:In cases where it is infeasible to compute the likelihood exac
 tly\, an  alternative is to find some numerical approximation to the likel
 ihood\,  then to use this approximate likelihood in place of the exact lik
 elihood  to do inference about the model parameters. This is a fairly comm
 only  used approach\, and I will give several examples of approximate  lik
 elihoods which have been used in this way. But is this a valid  approach t
 o inference? I will give conditions under which inference with  an approxi
 mate likelihood shares some of the same asymptotic properties  as inferenc
 e with the exact likelihood\, and describe the implications  in some examp
 les. I will finish with some ideas about how to construct  scalable likeli
 hood approximations which give statistically valid  inference.<br>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR