Actuarial Science and Insurance Education and events

Abstract:
After an brief introduction of the Statistics and Data Science group at  Department of Mathematics, University of Bergen, I will give a talk on  paper: Investigating the Asymmetric Behavior of Oil Price Volatility Using Support Vector Regression.

This paper investigates the asymmetric behavior of  daily or weekly oil price volatility from March 8, 1991 to September 13, 2019. Different types of Asymmetric Power ARCH (APARCH) model are implemented, and we compare the estimation and forecasting performance of the models estimated from the maximum likelihood estimation (MLE) method and support vector machine (SVM) based regressions. Combining nonparametric SVM method with parametric APARCH model not only enables to keep interpretations of the parametric models but also leads to more precise estimation and forecasting results. This paper, therefore, examines volatility behavior of oil price with both methodological and economic underpinnings.

Biography:
Yushu Li (Yushu Li | University of Bergen (uib.no)) works an associate professor in Statistics at Department of Mathematics, University of Bergen, Norway. In 2006, she obtained a bachelor's degree in Economics (major: Statistics, with direction in Actuarial Science) from School of Statistics, Renmin University of China. In 2011, she completed her PhD in Statistics and Econometrics at Linnaeus University in Sweden. She has worked as researcher at Lund University, Sweden and thereafter assistant professor at Norwegian School of Economics (NHH), before her position at University of Bergen from 2016.

Yushu Li's has worked on topics in Econometrics, especially nonlinear and nonstationary Time series Econometrics; Wavelet analysis in econometrics, and Statistical Surveillance. Her current research interest includes Sparse Bayesian Learning and Statistical machine learning, especially in prediction and evaluation of prediction.

To attend, please register here.

"Order statistics in natural catastrophe risk modelling”

Abstract

We start by discussing how natural catastrophe risk models are used in the re/insurance value chain, and recent work to quantify the effects of climate change. We then review various commonly discussed loss perspectives (AEP, OEP, CEP, EEF), before providing the definition of the order statistics, making clear their importance in re/insurance. Next, a proof of the general formula for the order statistics is provided, after which we discuss the application of the formula to the canonical frequency distribution  (the combination of the Negative Binomial, Poisson and Binomial). We then visualize how the order statistic exceedance probabilities change under alternative model assumptions, yielding insights as to how reinsurance pricing metrics may change under alternative climates. Finally, we review experiments with a toy catastrophe model which yields insights into the role that the order statistics play in reinsurance pricing metrics. This talk is based on two recent peer reviewed publications which will be referenced in this talk”.

Biography

Shree Khare is Vice President of model development for Moody’s RMS, and head of physical risk modelling under climate change. Shree currently leads a new initiative to integrate RMS models into applications for the entire financial sector. He began his career in risk modelling in 2006 with RMS, and he also spent 3 years working for Hiscox as the Group Head of Catastrophe Research, responsible for the view of risk applied across the business. Shree has contributed to (both as project lead and individual contributor) to various RMS projects and products (Japan, Taiwan and Korea typhoon; Catastrophe Response; North American Hurricane; European Windstorm; uncertainty quantification and clustering). Shree has an academic background in Physics (BSc ,University of British Columbia), Financial Mathematics (MSc, University of York), and Atmospheric Modelling and Prediction (PhD, Princeton University). Shree has made original research contributions to various academic fields including data assimilation, natural catastrophe risk modelling, reinsurance pricing and statistics.

"Pricing Insurance Risk: Theory and Practice"

Abstract:

This presentation discusses the main results from my monograph with John Major published earlier this year. It focuses on portfolio pricing and its allocation to individual units. We show how the standard assumption of a constant cost of capital (which goes back to Adam Smith) requires two risk measures, not one, and is inconsistent with observed bond credit spreads and cat bond pricing. We then explain how spectral risk measures provide a theoretically satisfying pricing model, but one that doesn't hold in practice because of subtle non-differentiability problems, which manifest themselves to practitioners as the "order problem." The theoretical model is also hard to calibrate. Next, we invert the calibration problem and provide a concrete description of the set of spectral risk measures that determine the same price for a given portfolio. We show how pricing functionals in this set can be interpreted as tail-centric or volatility-centric. They give different allocations to a net sub-portfolio over a range material to real-world decision-making. These findings are interpreted in light of current catastrophe reinsurance market dislocations.

Biography

Dr. Stephen Mildenhall is Faculty and Director of Insurance Data Analytics in the School of Risk Management, Insurance and Actuarial Science at St. John’s University’s Peter J. Tobin College of Business.

Dr. Mildenhall was Global CEO of Analytics for Aon and head of Aon Benfield Analytics where he helped found and establish Aon’s Singapore Center for Innovation and Analytics and led a team of over 500 professionals in actuarial science, catastrophe modeling, accounting and financial modeling. Prior to Aon, he was vice president of Actuarial Pricing for Kemper Insurance and began his career at CNA.

Dr. Mildenhall is a frequent speaker and published author on risk theory, the intersection of insurance and finance, and the application of probability and statistics to reserving and rate making problems. He is a fellow and Board and Audit Committee member of Casualty Actuarial Society (CAS), a board member of the CAS Institute, an associate of the Society of Actuaries, and a member of the Risk Theory Society. He served on the ARIA Board of Directors from 2013-2016 and was a member of the CAS Committee on the Theory of Risk and its chairman from 2008-2010.

He holds a B.Sc. in Mathematics from the University of Warwick, England, and a MA and Ph.D. in Mathematics from the University of Chicago. He is a Chartered Enterprise Risk Analyst, Certified Specialist in Predictive Analytics, and a Catastrophe Risk Management Professional.

His book, Pricing Insurance Risk: Theory and Practice, with John Major, was published by Wiley in 2022.