The Department of Mathematics is committed to producing world-class research, providing high quality undergraduate, graduate and professional programs of study that attract the best students, and to attending to the mathematical needs of the University and the community.

Henry R. Monteith Building, home of the Department of Mathematics.

- Liang Xiao awarded NSF CAREER grant
- “Complex Math Visuals are This Researcher’s Handiwork” — work of David Nichols, graduate student in mathematics, profiled by UConn Today
- Students present at 2018 Spring Frontiers Exhibition
- Actuarial program named Center for Actuarial Excellence for eighth consecutive year
- Graduate student receives CETL Outstanding Teaching Award
- Damir Dzhafarov spotlighted by the Connecticut Institute for the Brain and Cognitive Science
- Lan-Hsuan Huang Huang and Damin Wu receive appointments at IAS; Prof. Huang awarded Simons and von Neumann Fellowships
- “Very Special Snowflakes” — the work of Vyron Vellis (Assistant Research Professor in Math) featured in UConn Today
- Talitha M. Washington, first African American to receive math PhD from UConn, writes about journey from student to math professor, in the AMS Notices
- Neag School of Education, Office of the Provost, and the Department of Mathematics Present: Rachel Gutiérrez on “Rehumanizing Mathematics: Should That Be Our Goal?”

Connect:

+ Show/hide abstracts and details

Actuarial Science Seminar

Title: Application of Bayesian sensitivity analysis in compound risk model with random effects

Speaker: Himchan Jeong (University of Connecticut)

Time: Tuesday, December 11, 2018 at 11:00 am

Place: MONT 214Abstract: The generalized linear model (GLM) is a statistical model which has been widely used in actuarial practice, especially for insurance ratemaking. Due to the inherent longitudinal property of P&C insurance claim dataset, there have been some trials of incorporating unobserved heterogeneity of each policyholder from the repeated measurements. To achieve this goal, random effects model has been proposed but there was less theoretical discussion on the methods to test the presence of random effects in GLM framework. In this article, the concept of Bregman divergence is explored, which has some good properties for statistical modeling and can be connected to diverse model selection diagnostics as in Goh and Dey (2014). We can apply model diagnostics derived from the Bregman divergence for testing robustness of priors both on the naive model, which assumes that random effect has point mass as its prior density, and the proposed model, which assumes a continuous prior density of random effect. This approach provides the insurance companies concrete framework for testing the presence of random effects in both claim frequency and severity and furthermore appropriate hierarchical model which can explain both observed and unobserved heterogeneity of the policyholders for insurance ratemaking. Both models are calibrated using a claim dataset from the Wisconsin Local Government Property Insurance Fund which includes both observed claim counts and amounts from a portfolio of policyholders.

Title: Application of Bayesian sensitivity analysis in compound risk model with random effects

Speaker: Himchan Jeong (University of Connecticut)

Time: Tuesday, December 11, 2018 at 11:00 am

Place: MONT 214Abstract: The generalized linear model (GLM) is a statistical model which has been widely used in actuarial practice, especially for insurance ratemaking. Due to the inherent longitudinal property of P&C insurance claim dataset, there have been some trials of incorporating unobserved heterogeneity of each policyholder from the repeated measurements. To achieve this goal, random effects model has been proposed but there was less theoretical discussion on the methods to test the presence of random effects in GLM framework. In this article, the concept of Bregman divergence is explored, which has some good properties for statistical modeling and can be connected to diverse model selection diagnostics as in Goh and Dey (2014). We can apply model diagnostics derived from the Bregman divergence for testing robustness of priors both on the naive model, which assumes that random effect has point mass as its prior density, and the proposed model, which assumes a continuous prior density of random effect. This approach provides the insurance companies concrete framework for testing the presence of random effects in both claim frequency and severity and furthermore appropriate hierarchical model which can explain both observed and unobserved heterogeneity of the policyholders for insurance ratemaking. Both models are calibrated using a claim dataset from the Wisconsin Local Government Property Insurance Fund which includes both observed claim counts and amounts from a portfolio of policyholders.