Non-linear expectations of random sets and their application in statistics and utility theory (joint with Anja Muehlemann)
Date: 03/28/2019 (Thu)
Time: 3:30pm- 5:00pm
Location: Seminar will be held on-site: Social Sciences room 113
Organizer: Adam Rosen
Meeting Schedule: Login or email the organizer to schedule a meeting.
All meetings will be held in the same location as the seminar unless otherwise noted.
12:00pm - Lunch with Takuya, Francesca
1:30pm - Andrew Patton @228F
2:00pm - Federico Bugni @ 240
3:30pm - Seminar Presentation (3:30pm to 5:00pm)
6:00pm - Dinner with Adam, Fede, Francesca, Jia ...
Additional Comments: Abstract: Random sets are nowadays used in econometrics to describe and analyse partially identified models. One of standard tools is the concept of expectation of a random set. Motivated by recent advances in the theory of sublinear expectations and their applications in risk measurement and utility theory, the talk elaborates the concepts of sublinear and superlinear expectations defined on random closed sets. These expectations are set-valued and, unlike expectations of random variables, cannot be transformed to each other by altering the sign. Several general construction methods for non-linear expectations are presented. It is explained that sublinear expectations are naturally related to detecting outliers in samples of random sets, while superlinear ones can be used to assess multivariate utilities.