BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/68
DTSTAMP:20230914T125909Z
SUMMARY:Incorporating Views in Mathematical Models: An Approach Based on En
 tropy
DESCRIPTION:Speaker: Santanu Dey\nSchool of Technology and Computer Science
 \nTata Institute of Fundamental Research\nHomi Bhabha Road\n\nAbstract: \n
 A mathematical model based on historical data or general past experience m
 ay at times be an unsatisfactory model for the future. One way to come up 
 with a more accurate model is to explicitly incorporate in it views that a
 re believed to better reflect the future. We address this issue by letting
  $\\\\mu$ denote the original probability measure of a mathematical model.
  We then search for a probability measure $\\\\nu$ that minimizes a distan
 ce measure with respect to $\\\\mu$ and satisfies certain user specified v
 iews or constraints. We consider Kullbach-Leibler distance as well as othe
 r $f$-divergences as measures of distance between probability measures. We
  show that under the KL distance\, our optimization problem may lack a clo
 sed form solution when views involve fat tailed distributions. This drawba
 ck may be corrected if a ``polynomial-divergence  is used. We also discuss
  the optimal solution structure under these divergences when the views inc
 lude those on marginal probabilities associated with the original probabil
 ity measure. We apply these results to the area of portfolio optimization 
 where we note that a reasonable view that a particular portfolio of assets
  has heavy tailed losses leads to a more realistic model through our appro
 ach.\n
URL:https://www.tcs.tifr.res.in/web/events/68
DTSTART;VALUE=DATE:20100209
LOCATION:A-212 (STCS Seminar Room)
END:VEVENT
END:VCALENDAR