Department of Mathematics Calendar

Back to Listing

The Department of Mathematics Calendar has moved to Webtools. Anyone may submit an event by clicking the "+" button (upper right). Note: the Sponsor field is required. Just type "n/a"

All events will be reviewed before acceptance. Please email Shelby Koehne if you have any questions about submitting an event.

Note: you may search past and future events by clicking on the magnifying glass icon on the main Calendar page.

For an archive of past events: https://math.illinois.edu/research/seminars-department-calendar

Fall 2021 Actuarial Science and Financial Mathematics Seminar: Comparison of Model Selection Information Criteria for Distinguishing Operational Risk Models

Event Type
Seminar/Symposium
Sponsor
n/a
Location
https://illinois.zoom.us/j/7922219559?pwd=OFk4d0FwMWFNTnlIdXFJbytaN09Qdz09
Virtual
wifi event
Date
Dec 17, 2021   9:30 - 10:30 am  
Speaker
Daoping Yu, Univ. of Central Missouri
Contact
Frank Quan
E-Mail
zquan@illinois.edu
Views
11

(If there is a problem with the URL, please try to join the meeting with Meeting ID and Password.)

Abstract: We consider three model selection information criteria to assess the operational risk models. They are Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Information Complexity (ICOMP) based Criterion. We compare the performances of the three information criteria on distinguishing the operational risk severity models. The competing models are constructed using Champernowne, Frechet, Lognormal, Lomax, Paralogistic, and Weibull distributions, respectively. We conduct both simulation studies and a case study. It seems that no single information criterion is absolutely more effective than others in the simulation studies. In the case study, certain distributional models conform to the external fraud type of operational losses data in retail banking of Chinese banks. However, those models are difficult to distinguish using standard information criteria such as AIC and BIC. We have found the ICOMP criterion conveys a little bit more information when AIC and/or BIC cannot separate the Lognormal model out of the pool of competing models.

About: Daoping Yu received his Ph.D. degree in Mathematics with concentration in Statistics and Actuarial Science after obtaining his M.S. degree in Mathematics and B.S. degree in Information Management and Information Systems. He joined the School of Computer science and Mathematics at the University of Central Missouri in 2016. He has taught a variety of courses in mathematics, statistics and actuarial science. The courses include but are not limited to Statistical Modeling (STAM), Life Contingencies Models (LTAM), Mathematical Statistics (VEE), Review for Exams Probability (P) and Financial Mathematics (FM), Calculus, and Basic Statistics. His research is in the interface of actuarial science and financial mathematics with particular emphasis on the statistical modeling of operational risk management. He is an Associate of the Society of Actuaries (ASA).

link for robots only