BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:What is the chance that the match is coincidence? - Richard Gill\,
  Leiden University
DTSTART:20130524T150000Z
DTEND:20130524T160000Z
UID:TALK45268@talks.cam.ac.uk
CONTACT:Richard Samworth
DESCRIPTION:I'd like to talk about two topics both connected to forensic s
 tatistics on\nwhich I have been working recently. The 2010 UK Court of App
 eal Ruling known\nas "R v T" asserted that Bayes' theorem and likelihood r
 atios should not be\nused in evaluating forensic evidence\, except for DNA
  and 'possibly other\nareas where there is a firm statistical base'. This 
 illustrates that the\ntask of communicating the evidential value of statis
 tical evidence to a\ncourt is not easy.\n\nThe first\, more mathematical\,
  main topic\, is the subject of an ongoing\ncollaboration with Dragi Anevs
 ki (Lund). Think of it as a nonparametric\nmissing data estimation problem
  with parameter restricted by ordering\nconstraints. \n\nConsider a probab
 ility distribution over an infinite set. Let p=(p_1\,p_2\,?)\nbe the vecto
 r of all the atoms of this probability distribution ordered from\nlarge to
  small and augmented with zero's if there are only finitely many\natoms. N
 ow take an iid sample of size n from this distribution\, count how\noften 
 each element is observed\, and similarly order the resulting counts\nfrom 
 large to small. This results in our observed data Y=(Y_1\,Y_2\,?). The\npr
 oblem we are interested in is how to estimate the underlying vector of\nor
 dered probabilities p from the observed vector of ordered counts Y. Note\n
 that the k'th most frequent element in the sample is not necessarily the\n
 k'th most frequent element in the population!\n\nI will discuss our prelim
 inary results on the nonparametric maximum\nlikelihood estimator of p\, wh
 ich is very different from the naieve estimator\nY/n\, and explain their r
 elevance to the problem of evaluating the evidential\nvalue of a rare Y-ch
 romosome match - a problem called "the fundamental\nproblem of forensic ma
 thematics" by Charles Brenner (Berkeley).\n\nThe second topic concerns the
  problem of deciding whether or not two mobile\ntelephones actually belong
  to the same person\, based on call records (times\nof calls\, locations o
 f GSM towers) of both of the phones. Here there is no\nsimple model and no
  simple mathematical problem to be solved\, but on the\nother hand an equa
 lly challenging problem of how the statistician can advise\na court on the
  evidential value of the evidence. The court in question will\nbe the Unit
 ed Nations Special Tribunal on Lebanon\, the crime is the\nassassination o
 f premier Rafiq Hariri in 2005.\n
LOCATION:MR12\,  Centre for Mathematical Sciences\, Wilberforce Road\, Cam
 bridge
END:VEVENT
END:VCALENDAR