BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Talks.cam//talks.cam.ac.uk// X-WR-CALNAME:Talks.cam BEGIN:VEVENT SUMMARY:Optimal Design of Experiments on Connected Units with Application to Social Networks - Ben Parker (University of Southampton) DTSTART:20160826T123000Z DTEND:20160826T131000Z UID:TALK67072@talks.cam.ac.uk CONTACT:INI IT DESCRIPTION:Co-authors: Steven G. Gilmour and John Schormans< /span>  \;

When experiments are performed on social networks\, it is difficult to justify the usual assumpt ion of treatment-unit additivity\, due to the connections between actors i n the network. We investigate how connections between experimental units affect the design of experiments on those experimental units. Specificall y\, where we have unstructured treatments\, whose effects propagate accor ding to a linear network effects model which we introduce\, we show that optimal designs are no longer necessarily balanced\; we further demonstra te how experiments which do not take a network effect into account can lea d to much higher variance than necessary and/or a large bias. We show the use of this methodology in a very wide range of experiments in agricultu ral trials\, and crossover trials\, as well as experiments on conne cted individuals in a social network. LOCATION:Seminar Room 1\, Newton Institute END:VEVENT END:VCALENDAR