BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Bayesian Unfolding Method Applied to the PAMELA Experiment - Ger
 mano Percossi
DTSTART:20060620T130000Z
DTEND:20060620T140000Z
UID:TALK5056@talks.cam.ac.uk
CONTACT:David MacKay
DESCRIPTION:PAMELA is a satellite-borne experiment that is going to study 
 cosmic\nrays in a wide energy range and for a very long period (approx. 3 
 yrs)\,\nwith an unprecedented precision.\nIts main scientific objectives a
 re the indirect study of possible dark\nmatter candidates and the search f
 or antiparticles coming directly from\nantimatter domains.\n\nThese scient
 ific tasks require a good spectrum reconstruction and\nparticle separation
  in order to appreciate weak signals over\na strong background (e.g positr
 ons over protons) that could be hints for\nthe phenomena PAMELA is going t
 o look for.\nFor inference problems like these\, we think the Bayesian sta
 tistics (or\n"probabilistic approach") offers a better  tool than standard
 \nstatistics\, both conceptually and practically.\n\nWe present some resul
 ts\, based on Montecarlo simulations\, that\nshows how we applied an unfol
 ding algorithm (D'Agostini 1995) to\nreconstruct a positrons spectrum over
  a stronger protons background.\nWe used only observables of a subset of P
 AMELA's\, obtaining good results\n anyway.\nWe show also how we solved som
 e practical problems we faced with during\nour work.\n
LOCATION:HEP Seminar Room\, Cavendish Laboratory
END:VEVENT
END:VCALENDAR