BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:An adaptive optimal design with a small fixed stag
 e-one sample size - Flournoy\, N (University of Mi
 ssouri-Columbia)
DTSTART;TZID=Europe/London:20150706T133000
DTEND;TZID=Europe/London:20150706T141500
UID:TALK60047AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/60047
DESCRIPTION:A large number of experiments in clinical trials\,
  biology\, biochemistry\, etc. are\, out of necess
 ity\, conducted in two stages. A first-stage exper
 iment (a pilot study) is often used to gain inform
 ation about feasibility of the experiment or to pr
 ovide preliminary data for grant applications. We 
 study the theoretical statistical implications of 
 using a small sample of data (1) to design the sec
 ond stage experiment and (2) in combination with t
 he second-stage data for data analysis. To illumin
 ate the issues\, we consider an experiment under a
  non-linear regression model with normal errors. W
 e show how the dependency between data in the diff
 erent stages affects the distribution of parameter
  estimates when the first-stage sample size is fix
 ed and finite\; letting the second stage sample si
 ze go to infinity\, maximum likelihood estimates a
 re found to have a mixed normal distribution.\n
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:
END:VEVENT
END:VCALENDAR