BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Experience and challenges of applying Bayesian methods in food saf
 ety risk assessments - Dr. Marc Kennedy\, The Food and Environment Researc
 h Agency (Fera)\, Risk and Numerical Sciences (RANS) team
DTSTART:20110517T133000Z
DTEND:20110517T143000Z
UID:TALK30832@talks.cam.ac.uk
CONTACT:Li Su
DESCRIPTION:The RANS team at Fera works with various government agencies\,
  regulatory bodies and software developers in the UK and EU to try and dev
 elop and apply improved methods for probabilistic risk assessments across 
 a range of policy areas. My work has been focused on quantifying uncertain
 ty and variability within risk assessment for dietary exposure to pesticid
 es. For risk managers it is clearly important to separate uncertainty from
  variability\, so they can decide whether they should be collecting more d
 ata to reduce uncertainty or taking steps to reduce inherent variability b
 y changing the system.\n\n \n\n2-dimensional Monte Carlo (2DMC) algorithms
  have been used as the standard approach to separately represent uncertain
 ty and variability\, but these are frequently used incorrectly. Lognormal 
 distributions are often assumed for the variability distribution as standa
 rd without justification. Also\, many sources of uncertainty are ignored\,
  or treated using ad-hoc methods. Problems include:\n\n \n\n·         Lim
 ited residue measurements from random market surveys\n\n·         Extreme
 ly high proportions of non-detects or missing residues\n\n·         Compo
 site measurements from a small set of food units treated as a batch/field 
 mean residue\n\n·         Measurement uncertainty of residues\n\n·      
    Limited dietary survey used to infer whole population eating habits (ov
 er a lifetime)\n\n·         Pesticide residues can come from multiple sou
 rces\, e.g. a child might eat an apple and a pear on any given day\, both 
 of which were treated with pesticides\n\n·         Lognormal often fits b
 adly\, but what is the 'right' distribution?\n\n \n\nI will talk about how
  Bayesian methods have been applied in a range of applications and list so
 me open research problems. Computational aspects will also be mentioned.\n
 \n \n\n 
LOCATION:Large Seminar Room\, 1st Floor\, Institute of Public Health\, Uni
 versity Forvie Site\, Robinson Way\, Cambridge
END:VEVENT
END:VCALENDAR