BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Scott Sheffield (MIT) DTSTART:20200529T150000Z DTEND:20200529T160000Z DTSTAMP:20260423T005721Z UID:PatC/6 DESCRIPTION:Title: Pro bability and pandemics\nby Scott Sheffield (MIT) as part of Probabilit y and the City Seminar\n\n\nAbstract\nIn one of the simplest epidemic mode ls\, one lets $p_n$ denote the number of new infections during week $n$ an d assumes that (during the early stages of the epidemic) $p_{n+1} = R_0 p_ n c_n$ where $c_n$ measures the "fraction of usual contact" that takes pla ce between people during the nth week. Within this simplistic model\, inte rmittent strategies (taking $c_n$ small some weeks and large other weeks) lead to lower infection rates than consistent strategies with the same tot al amount of contact.\n\nBut what happens if one considers a more realisti c disease model (such as a SEIR model with multiple compartments\, or a ne twork-based model\, with empirically based distributions for incubation an d infection times) and also tries to assign utility to the amount of conta ct in a more realistic way (accounting for crowding\, social networking an d other issues)? What factors cause intermittent strategies to outperform constant strategies? I will discuss a health policy paper I recently co-au thored with a team of public health researchers that explores this questio n for a range of simple examples.\n LOCATION:https://researchseminars.org/talk/PatC/6/ END:VEVENT END:VCALENDAR