MCFAM Seminars

Speaker: Lukas Gonon

Abstract: In this talk we present a novel deep learning methodology to detect financial asset bubbles by using observed call option prices. The proposed algorithm is widely applicable and model-independent. We test the accuracy of our methodology in numerical experiments within a wide range of models and apply it to market data of tech stocks in order to assess if asset price bubbles are present. In addition, we provide a theoretical foundation of our approach in the framework of local volatility models. To this purpose, we give a new necessary and sufficient condition for a process with time-dependent local volatility function to be a strict local martingale.

Joint work with Francesca Biagini, Andrea Mazzon and Thilo Meyer-Brandis.

 Bio: Lukas is a Senior Lecturer (UK equivalent to US assistant professor) jointly appointed at the Department of Mathematics of Imperial College London and the artificial intelligence initiative Imperial-X (I-X). Lukas holds a doctoral degree from ETH Zürich. Before joining Imperial College London he was also an assistant professor at University of Munich. His research is at the intersection of mathematics, machine learning and finance. It centres around various machine learning methods (deep learning, reservoir computing, random features, kernel methods, ...) and their applications to time series, stochastic processes, partial differential equations and finance.

Speaker: Artur Sepp

Abstract: We introduce the log-normal stochastic volatility (SV) model with a quadratic drift to allow arbitrage-free valuation of options on assets under money-market account and inverse martingale measures. We show that the proposed volatility process has a unique strong solution, despite non-Lipschitz quadratic drift, and we establish the corresponding Feynman–Kac partial differential equation (PDE) for computation of conditional expectations under this SV model. We develop an analytic approach to compute an affine expansion for the moment generating function of the log-price, its quadratic variance (QV) and the instantaneous volatility. Our solution allows for semi-analytic valuation of vanilla options under log-normal SV models closing a gap in existing studies. We apply our approach for solving the joint valuation problem of vanilla and inverse options, which are popular in the cryptocurrency option markets. We demonstrate the accuracy of our solution for valuation of vanilla and inverse options. By calibrating the model to time series of options on Bitcoin over the past four years, we show that the log-normal SV model can work efficiently in different market regimes.

Bio: Artur Sepp is the head quant at LGT bank in Zurich focusing on quantitative asset allocation and systematic investment strategies. Artur has over 15 years of experience in financial markets including heading quant research and portfolio management at a systematic hedge fund and a family office as well as leading development of front-office quant strategies and derivatives at private (Julius Baer) and investment banks (Merrill Lynch/BofA). Artur has a PhD in Mathematical Statistics from University of Tartu, an MSc in Industrial Engineering and Management Sciences from Northwestern University, and a BA cum laude in Mathematical Economics from Tallinn University of Technology. His expertise covers quantitative investing and asset allocation, quantitative modelling of derivative securities, machine learning and data science, and blockchain applications within decentralised finance. He is the author and co-author of several research articles on quantitative finance published in key journals. Artur won Risk Magazine’s Quant of the Year Award (2024).

Speaker: Viatcheslav Belyaev

Abstract: Local Volatility (LV) is a very powerful tool for market modeling. This tool can be used to generate arbitrage-free scenarios calibrated to all available options. Here we demonstrate how to implement LV in order to reproduce most swaption prices within a single model. There was a good agreement between market prices and Monte Carlo prices for all tenors and maturities from 2 to 20 years.

Bio: Viatcheslav Belyaev received his PhD degree from the Institute of Theoretical and Experimental Physics (Moscow, Russia). He worked as Quantitative Analyst in Allianz Life during 2005-2018, then and currently in U.S. Bank. His activity is concentrated on modeling and numerical methods for interest rates, equity, cross currency, credit and XVA and its applications. VB is an author of multiple publications in mathematical finance and a frequent speaker at QuantMinds International Conferences.
 

February 16, 2024

Speaker: Xianhui Lei (Directed by Prof. Arkady Shemyakin)

Abstract: This study evaluates COVID-19 mortality in Minnesota and Wisconsin using time series analysis and copulas to model the pandemic's impact from an actuarial standpoint. By applying ARIMA methods to weekly mortality data and analyzing residuals, it identifies deviations from expected distributions. The research employs pair copulas to explore statistical dependencies between states, recommending the Student t-copula for ARIMA residuals. This approach offers precise short-term mortality projections, contributing to epidemiological forecasting and public health planning.

Bio: Xianhui Lei is a second year MFM student at UMN with a robust foundation on actuarial science and statistics, interested in applying mathematical models to explore the dynamics of financial markets.

Zoom Recording

February 23, 2024 

Speaker: Thorsten Moenig

Abstract: Many Americans continue to be financially underprepared for their retirement. Automatic enrollment in employer-sponsored qualified retirement plans (QRPs) has helped, but the target-date funds commonly used as default investment choices have their own problems. We develop a target-date version of the recently developed Registered Index-Linked Annuities (RILAs) and suggest that these ``TD-RILAs'' can provide a more cost-effective and more transparent way to attain a diversified equity exposure, the level of which decreases over time. A theoretical analysis explores the optimal structure of TD-RILAs and their comparison to target-date funds from an investor's perspective. A large-scale lab experiment sheds further light on investors' preferences, focusing on the importance of product transparency, of the employer's default investment choice, and of the role of information in improving financial literacy. We conclude that TD-RILAs would be a suitable addition to QRPs and may even rival target-date funds as Qualified Default Investment Alternatives.

We make three primary contributions: First, we introduce a novel investment product as an alternative to target-date funds by embedding RILAs with annual crediting periods in a ``glide path'' structure. Second, we expand the literature on RILAs by showcasing a new application for this emerging product. This also adds to the broader literature on retirement savings products and strategies, in view of ongoing challenges with financial literacy. And third, our experiment enriches the behavioral retirement literature, emphasizing the crucial role of personalized information in influencing investment decisions.

Bio: Thorsten Moenig is an Associate Professor in the Department of Risk, Actuarial Science & Legal Studies in the Fox School of Business at Temple University, and the director of the Actuarial Science Masters’ program. His current research focuses on personal retirement savings products (such as variable annuities and RILAs), how insurers can value and hedge the embedded guarantees, and how these products should optimally be designed in view of policyholder behavior and market imperfections. More broadly, he is interested in topics across actuarial science, insurance economics, applied game theory, and behavioral economics. Dr. Moenig has published his research in Insurance: Mathematics and Economics, the Journal of Risk and Insurance, the North American Actuarial Journal, and the Review of Finance. He is a recipient of the 2017 Redington Prize from the Society of Actuaries and serves on the Editorial Board of the Journal of Risk and Insurance.

Dr. Moenig holds a Master’s degree in Actuarial Science from the University of Connecticut and a PhD in Risk Management and Insurance from Georgia State University. He is an Associate of the Society of Actuaries (ASA). He enjoys teaching and is the proud recipient of two Outstanding Teacher Awards by the Sigma chapter of Gamma Iota Sigma.

Zoom Recording

Presentation

March 1, 2024 

Speaker: Gary Hatfield

Abstract: Quantitative Finance is a dynamic and highly interdisciplinary field that offers a variety of highly rewarding career opportunities. If you are majoring in Mathematics, Statistics, Actuarial Science, Computer Science, Economics, Finance, or another STEM field, you are likely a good candidate for this career. This seminar will provide valuable insights into the profession, discussing the roles of quantitative analysts ("quants") and exploring potential pathways to enter this exciting field.

Join us to learn about this fulfilling career and for an in-depth discussion on one particular pathway - pursuing a Masters in Financial Mathematics (MFM) at the University of Minnesota. The seminar will also feature two early-career financial quants who will be joining to add their perspectives and share their experiences.

Bio: Gary Hatfield is affiliated with the Minnesota Center for Financial and Actuarial Mathematics (MCFAM) at the University of Minnesota. Prior to his full-time role with MCFAM, Gary spent over 20 years in industry as an Actuary and a Quant working with Securian. 

At Securian, he played a key role in building out functions for hedging, ALM, ERM, and Data Analytics. He was the lead developer of the Society of Actuaries’ Quantitative Finance and Investments (QFI) Fellowship Track and one the founding practitioners who developed the University’s

Zoom Recording

Abstract: These former students have landed rewarding careers in Quantitative Finance, and they're eager to share their job search journeys with you.

Zoom Recording

April 5, 2024 

Speaker: Liwei Huang - MFM Alumnus -  AIM Ph.D. - University of Michigan

Abstract: In this talk, we present a trading game involving strategic liquidity providers under different inventory risks aversion. The liquidity providers compete to supply liquidity to a risk-averse agent who is privately informed about the final value of asset and her own initial endowment. We show that there exists a unique asymmetric Nash equilibrium with convex schedules, and it can be characterized by the solution of the system of quasimonotone ODEs. This is based on a joint ongoing work with Ibrahim Ekren.

Bio: Liwei Huang holds an undergraduate degree in financial mathematics from Henan University of Economics and Law. He further pursued his education at the University of Minnesota, where he completed his Master’s of Financial Mathematics (MFM) in Spring 2020. Liwei has since continued to advance his academic journey and is currently an Applied and Interdisciplinary Mathematics (AIM) PhD student within the Mathematics Department at the University of Michigan. His research primarily focuses on dynamic principal-agency models and Stackelberg Equilibrium.

Zoom Recording

April 12, 2024 

Speaker: Deqing Jiang

April 19, 2024

Speaker: Benjamin Hood

Abstract:  Why do volatility targeting/management strategies tend to outperform simple buy-and-hold positions in the same assets, as found by Moreira and Muir (2017) and Harvey et al (2018)? We test the hypothesis that this outperformance is mainly due to a loading on trend following that arises because of the negative correlation between return direction (trend) and magnitude (volatility), the so-called “leverage effect.” When controlling for trend exposure, alpha to volatility targeting is shown to mainly accrue to trend for both a long equity history and a set of 14 global equity index futures contracts. By contrast, this is not true for commodity, fixed income, or currency futures, where the leverage effect is not present. We further discuss the mechanical relationship between volatility targeting and trend following, creating a point of connection between these two seemingly different branches of research.

Bio:  Ben is Managing Director of Derivatives Research at Morgan Stanley and Parametric Portfolio Associates.  Research interests include alternative risk premia, volatility and risk modeling, and multi-asset portfolios.  Prior to joining Parametric, he worked as a vice president of research at AQR Capital Management.  Ben’s work has been published in the Review of Financial Studies, the Journal of Derivatives, and the Financial Analysts Journal.  He earned a PhD and MA in economics from UCLA, as well as a BS in economics with honors and BA in mathematics from the University of Minnesota.

Date: September 15, 2023 

Panelist:

Holly Ainslie: Holly Ainslie is Vice President at Ameriprise Financial, Inc. leading the team responsible for derivatives procedures and policy for product-based hedging in the Capital Markets Solutions (CMS) group. Prior to joining Ameriprise in 2007, she was with GMAC RESCAP, where she was an analyst in a group responsible for developing risk management tools to value, analyze and predict behavior of residential mortgages and other asset backed securities. She holds a Master of Financial Mathematics (University of Minnesota) and a BA in Economics, Quantitative Emphasis (University of Minnesota). 

Chris Prouty: Chris Prouty is a local quantitative finance practitioner working for Cargill, Inc.  He has spent most of his career in trading roles in the OTC commodity derivatives space in markets spanning agriculture, softs, and metals.  For the past few years, he has focused on building systems to support trading in OTC commodity derivatives.  He has served in the Master of Financial Mathematics program as an instructor since 2007, teaching the implementation of financial math models in modern programming languages.

Robert Schell: Robert Schell is a 2016 graduate of the Master of Financial Mathematics program from the University of Minnesota, with an undergraduate degree in Economics also from the “U”.  He has worked at U.S. Bank for 12 years in a variety of regulatory-change roles, including the LIBOR transition and Basel capital requirements for non-loan portfolios.  Robert recently moved to the new Competitive Intelligence function housed within U.S. Bank’s Investor Relations.  Separately, he contributes market and macroeconomic views to individual areas within the bank.

Zoom Recording

Date: September 29, 2023 

Speaker: John Wald, PhD.  

Abstract: We consider the problem of optimal dynamic trading in the presence of predictable returns and proportional transaction costs for an investor choosing among multiple assets.  The value of each security equals the expected value of holding the asset plus the value of all options to trade.  We provide exact trading rules for N-assets that follow an MA(1) process.  Simulations demonstrate the impact of transaction costs, volatility, and predictability on optimal trading behavior.  The optimal trading rule can substantially increase performance if transaction costs vary among assets.

Bio: John Wald is Professor of Finance at the Alvarez College of Business at the University of Texas at San Antonio.  He received his Ph.D. in economics from the University of California at Berkeley.  His current research interests include sustainable finance, corporate governance, and foreign exchange exposure.  He has also dabbled in models about trading.

Presentation

Date: October 6, 2023 

Speaker: Jean-François Bégin, PhD, FSA, FCIA

Abstract: Over the last 40 years, various frameworks have been proposed to model economic and financial variables relevant to actuaries. These frameworks—called economic scenario generators—are comprehensive models that allow actuaries and risk managers to grasp the long-term uncertainty underlying financial market values and economic variables. Their primary aim is to generate a set of future scenarios covering a range of plausible outcomes. The main end-users of these frameworks are pension, life insurance, and banking practitioners. In this presentation, we explore two important modelling questions relating to economic scenario generators and their use: model uncertainty and model averaging.

Given today’s knowledge and technology, one could construct complicated frameworks to fit the data better. However, this process would lead to highly parametrized models, which goes against the idea of parsimony in statistics—the desire to explain phenomena using fewer parameters. The first part of this presentation investigates this tradeoff: would a more complex economic scenario generator perform better, or would a simple model accomplish the same performance? To answer this question, we propose a new complex generator that nests versions of well-known actuarial frameworks. We then assess whether complex models perform better than simple models using both in- and out-of-sample analyses.

Second, we investigate model averaging. This strategy has been used extensively in data-heavy domains such as weather forecasting and in fields where forecasts come from diverse methods and datasets, such as election polls. In our context, we answer whether we can create better, more reliable economic scenario generators by combining them. This part of the presentation considers a recently proposed Bayesian averaging technique and data from multiple countries.

Bio: Dr. Jean-François Bégin is an Associate Professor of Actuarial Science in the Department of Statistics and Actuarial Science at Simon Fraser University in British Columbia, Canada. He is a specialist in financial modelling as well as statistical and mathematical applications to finance and insurance. Before joining Simon Fraser University, he received his PhD from HEC Montréal in Financial Engineering. He is also a Fellow of both the Society of Actuaries and the Canadian Institute of Actuaries.

Over the past few years, his research program focused on four main areas: the construction, estimation, and implementation of complex models for long-term economic predictions, the understanding and management of credit risk, the modelling of option pricing, their hedging, and their use in risk management, and the development of sustainable retirement solutions and designs.

Zoom Recording

Presentation
 

Date: October 27, 2023

Speaker: Anne MacKay, PhD. 

Abstract: We consider the pricing of variable annuities (VAs) with general fee structures under a class of stochastic volatility models. In particular, we analyze the impact of different VIX-linked fee structures on the optimal surrender strategy of a VA contract with guaranteed minimum maturity benefit (GMMB). We develop efficient algorithms for the pricing of VA contracts using a two-layer continuous-time Markov chain approximation for the fund value process. When the contract is kept until maturity and under a general fee structure, we show that the value of the contract

can be approximated by a closed-form matrix expression. We also provide a quick and simple way to determine the value of early surrenders via a recursive algorithm and give an easy procedure to approximate the optimal surrender surface. We show numerically that the optimal surrender strategy is more robust to changes in the volatility of the account value when the fee is linked to the VIX index.

Bio: Anne MacKay received a Ph.D. in Actuarial Science from the University of Waterloo, in 2014. From 2014 to 2016, she was a Postdoctoral Fellow at Risklab, ETH Zurich. From 2016 to 2021, she was an Assistant Professor, and then an Associate Professor at the Université du Québec à Montréal. Since 2021, she is an Associate Professor in the Department of Mathematics and Department of Finance at the Université de Sherbrooke. Her research interests include optimal stopping problems and numerical methods in finance and insurance.

Prof. MacKay has been a Fellow of the Society of Actuaries since 2012 and an Associate of the Canadian Institute of Actuaries since 2019.

Presentation
Presentation

Recording

Date: November 17, 2023 

Speakers: Nellie Garcia and Emily Gullerud - PhD. Program School of Mathematics University of Minnesota

Date: April 28, 2023 

Speaker: Tianze Li

Abstract: Estimating life expectancy is crucial in the life settlement industry, but traditional methods such as Proportional Hazard models can be inefficient when taking into consideration possible co-morbidities. In this seminar, we present a novel approach to life expectancy estimation by transforming the problem into a classification task using a tree-based ensemble model. This method is more efficient at considering co-morbidities and provides significantly better performance than traditional methods. Our approach has the potential to improve the accuracy of life expectancy estimates and facilitate better decision-making in the life settlement industry.

Date: April 21, 2023

Speaker: Indranil SenGupta

Abstract: In this presentation, we consider the portfolio optimization problem in a financial market under a general utility function. Empirical results suggest that if a significant market fluctuation occurs, invested wealth tends to have a notable change from its current value. We consider an incomplete stochastic volatility market model that is driven by both a Brownian motion and a jump process. We show a closed-form formula for an approximation to the optimal portfolio in a small-time horizon. We also prove the accuracy of the approximation formulas. Finally, we provide a procedure for generating a close-to-optimal portfolio for a finite time horizon.

Bio:  Indranil SenGupta received his Ph.D. in Mathematics from Texas A&M University in 2010. He is currently a Professor in the Department of Mathematics, and the mathematics graduate program director, at North Dakota State University. His research interests include Mathematical Finance, Stochastic Processes, and Data Science. He has published more than 50 papers/book chapters in various high-quality peer-review journals. His works appeared in notable journals such as SIAM Journal on Financial Mathematics, Annals of Finance, Applied Mathematical Finance, etc.  He has been invited to present his research in more than 70 national and international conferences/seminars/colloquiums. So far, he has been the research adviser for seven Ph.D. students. He is the winner of several teaching/service awards. He has served in the editorial board of various journals.  

Date: April 7, 2023

Speaker Bios: 

• Heeth Surana completed his Bachelor’s in Economics and Applied Mathematics from Macalester College, Minnesota. He is currently a first-year student in the MFM program at the University of Minnesota Twin Cities. 
• Abdullahi Abdullahi completed his Bachelor’s in Mechanical Engineering at the University of Minnesota and then went on to pursue a Masters in Financial Mathematics. 
• Erick Giron completed his Bachelor’s in Mathematics - Statistics at the California State University, Fresno and continues to pursue a Masters in Financial Mathematics at the University of Minnesota. 
• Brisseida Villarreal completed a Bachelor’s in Mathematics at the University of Texas Rio Grande Valley and has since them been pursuing a Masters in Financial Mathematics at the Univeristy of Minnesota. 

Abstract: Copulas are statistical methods for understanding the joint probabilities of a multivariate distribution, and have wide applications in quantitative finance to model tail risks. We have implemented a comparative study of pairs trading strategies on the NASDAQ 100 and the Russell 1000 Technology indices using a Gumbel and flipped (also termed survival) Gumbel copula. We divided 15 years (2008-2022) of data into two sections: 10 years of training (2008-2017) and 5 years of testing (2018-2022). We implemented the pairs trading strategy using the two copula approaches on the testing data. To grade the margins of the return data gathered from the two equity indices, we used the Friemer-Mudholkar-Kollia-Lin Generalized Lambda Distribution. A backtesting of the pairs strategy was executed on the testing data which showed that the flipped Gumbel copula outperforms the Gumbel copula. Our results showed that the flipped Gumbel copula approach for the pairs trading strategy traded with higher frequency and generated consistent alpha across all tested years while the Gumbel approach had a large negative alpha in the years 2019 and 2021. This implies that the indices exhibit greater correlations in their lower tails and are thus modeled better using the flipped Gumbel copula.

Zoom Recording

Bio: Frédéric Godin is an Associate Professor at the Mathematics and Statistics Department of Concordia University in Montreal, Quebec, Canada. His research interests are financial engineering, risk management, actuarial science, reinforcement learning, stochastic modeling, dynamics programing, variable annuities and energy markets. He holds the Fellow of the Society of Actuaries (FSA) and Fellow of the Canadian Institute of Actuaries (FCIA) designations. He am part of the Quantact research group.

Presentation

Zoom Recording

Date: February 24, 2023

Speaker: Eric Falkenstein 

Abstract: The most common blockchain trading contracts are based on a constant product rule, a*b=k, which subjects liquidity providers to negative convexity.  This ‘gamma cost’ for most contracts exceeds fee revenue. A solution is proposed to remove this cost by incenting liquidity providers to act as price-setting arbitrageurs while also preventing a subset of LPs from dominating the others.

Bio: Eric Falkenstein received his Ph.D. in economics from Northwestern in 1994 and wrote his dissertation on the low return to high volatility equities. He has written articles in several academic journals, including the Journal of Finance, and published two books—Finding Alpha and The Missing Risk Premium. He worked at KeyCorp as a risk manager and was head of economic capital allocations across the many bank business lines. At Moody's, he created their default model for private companies. He has been a long-short equity portfolio manager at various hedge funds and is currently working on Ethereum contracts and incubating a long-short equity strategy.

Zoom Recording

Date: February 17, 2023

Speaker: Adriana Ocejo, University of North Carolina

Abstract: We study a portfolio optimization problem involving the loss averse policyholder of a variable annuity with a guaranteed minimum maturity benefit. This financial guarantee is financed via a fee withdrawn directly from the investment account, which affects the net investment return. A fair pricing constraint is defined in terms of the risk-neutral value of the final contract payout. We propose a new fee structure that adjusts to the investment mix maximizing policyholder’s utility while keeping the contract fairly priced. We seek the investment strategy that maximizes the policyholder’s expected utility of terminal wealth after the application of a financial guarantee and subject to the fair pricing constraint. We assume that the policyholder’s risk attitude is relative to a reference level, risk seeking towards losses and risk-averse towards gains. We solve the associated constrained stochastic control problem using a martingale approach and analyze the impact of the fee structure on the optimal investment strategies and payoff. Numerical results show that it is possible to find an optimal portfolio for a wide range of fees, while keeping the contract fairly priced.

Recording

Date: December 2, 2022

Speaker: Paul Bilokon

Abstract: The principal component analysis (PCA) is a staple statistical and unsupervised machine learning technique in finance. The application of PCA in a financial setting is associated with several technical difficulties, such as numerical instability and nonstationarity. We attempt to resolve them by proposing two new variants of PCA: an iterated principal component analysis (IPCA) and an exponentially weighted moving principal component analysis (EWMPCA). Both variants rely on the Ogita–Aishima iteration as a crucial step.

Bio: Dr. Paul Bilokon is CEO and Founder of Thalesians Ltd and an academic at Imperial College, where his work focuses on machine learning, high performance computing, big data, and electronic trading. His career in quantitative finance spans Morgan Stanley, Lehman Brothers, Nomura, Citigroup, Deutsche Bank, and BNP Paribas, and he and his team at Thalesians continue to provide consulting services to numerous financial institutions, both on the buy-side and on the sell-side. He is one of the e-credit pioneers and has co-authored several books, including Machine Learning in Finance: From Theory to Practice and Big Data and High-Frequency Data with kdb+/q. Paul is fluent in C++, Java, Python, and kdb+/q and enjoys building distributed software systems powered by ML and applied mathematics.

Presentation

Recording

Date: November 18, 2022

Speaker: Dale Rosenthal

Abstract: The realized variance and high-frequency data literature has developed methods to correct for "microstructure noise." that noise is often thought to be due to bid-ask bounce. Using a variety of estimators for spreads and variance, I show that we can use these methods to also infer spreads. However, that reveals that the bias corrections (and thus the variance estimators themselves) are not what we think they are. This raises questions of what spreads and corrections we really want, what these methods are really estimating, and how we handle the endogeneity of trading.

Bio: Dale Rosenthal is a Director of Derivatives Research at Parametric, a Morgan Stanley company. He was previously a quantitative Eurodollars trader and consultant at his own firm, Q36; and, before that he was an assistant professor at UIC, UIUC, and Notre Dame. He interned at Goldman Sachs; worked for five years as an equity derivatives strategist at Long-Term Capital Management; and, was a quantitative researcher and proprietary algorithmic trader at Morgan Stanley's Equity Trading Lab. His academic research focuses on trading and financial distress using market microstructure and financial econometrics. His work has been published in the Journal of Financial Econometrics and the Journal of Empirical Finance. Since 2009 he has co-organized the R/Finance conference on using R for financial data analysis. He has appeared on TV and radio, been quoted in newspapers, and testified before the Illinois state legislature House Committee on Pension Investments. He has presented policy-relevant work at the European Central Bank, Banque de France, Austrian National Bank, National Bank of Slovakia, Bank of Finland, Reserve Bank of New Zealand, FDIC, and CFTC. Mr. Rosenthal graduated from Cornell University with a B.S. in electrical engineering and the University of Chicago with a Ph.D. in statistics.

Recording

Date: October 28, 2022

Speaker: Turan Bali, Georgetown University

Abstract:  We propose a theoretically motivated and empirically robust factor model for option returns. The model consists of factors based on option illiquidity, option price, implied-minus-realized volatility, implied-minus-realized skewness, implied-minus-realized kurtosis, and the option market factor. We find that the model has a higher tangent portfolio Sharpe ratio than extant factor models and outperforms such models at explaining the returns of a larger number of test assets. We also show that the stochastic discount factor implied by the newly proposed model explains the risk-return opportunities available in the options market and improves the cross-sectional pricing performance for individual equity options.

Recording

October 14, 2022

Speaker: Caroline Bang, Iowa State

Abstract: Unlike most fixed income securities, the timing and amount of mortgage-backed security cash flows is uncertain due to the fact that borrowers have the right to prepay the loan at any given time during the life of the loan. Prepayment of mortgages comes from five sources, one of which is turnover. Turnover is caused by moving and home upgrades, and it creates a baseline level of prepayment activity that is highly seasonal. Since it creates a baseline, understanding turnover is important to evaluating the risks of MBSs. We applied both a top-down approach using macroeconomic factors and a bottom-up approach using loan level data to gain a better understanding of turnover.

October 7, 2022

Speaker: Ping Zhong, University of Minnesota

Abstract: This work is undertaken through the MFM-directed study elective under the guidance of professor John Dodson. I am studying the static and dynamic nature of the equilibration process in a simple market consisting of agents trading complementary commodities in a sampled community graphon network. Interestingly, the convergence seems to be suboptimal over a wide range of graphon parameters and network sizes. It remains to be seen whether stochastic perturbation in the agent endowments or the network connections can lead to improved outcomes.

Zoom Recording

Date: April 22, 2022

Speaker: Lu Yang

Title: Dynamic Prediction of Outstanding Insurance Claims Using Joint Models for Longitudinal and Survival Outcomes

Abstract: To ensure the solvency and financial health of the insurance sector, it is vital to accurately predict the outstanding liabilities of insurance companies. We aim to develop a dynamic statistical model that allows insurers to leverage granular transaction data on individual claims into the prediction of outstanding claim payments. However, the dynamic prediction of an insurer's outstanding liability is challenging due to the complex data structure. The liability cash flow from a claim is generated by multiple stochastic processes: a recurrent event process describing the timing of the cash flow, a payment process generating the sequence of payment amounts, and a settlement process terminating both the recurrence and payment processes. We propose to use a copula-based marked point process to jointly model the three processes. Specifically, a counting process is employed to specify the recurrent event of payment transactions; the time-to-settlement outcome is treated as a terminal event for the counting process; and the longitudinal payment amounts are formulated as the marks associated with the counting process. The dependencies among the three components are induced using the method of pair copula constructions. Compared with existing joint models for longitudinal and time-to-event data such as random effect models, the proposed approach enjoys the benefits of flexibility, computational efficiency, and straightforward prediction.

Bio: Lu Yang is an Assistant Professor in the School of Statistics at the University of Minnesota. She received her Ph.D. in Statistics from the University of Wisconsin-Madison in 2017. Prior to joining UMN, she was an Assistant Professor in Actuarial Science and Mathematical Finance at the University of Amsterdam. Her current research focuses on multivariate analysis, nonparametric estimation of copulas, and regression model diagnostics, especially with discrete and semi-continuous outcomes.

Presentation

Zoom Recording

Date: April 1, 2022

Speaker: Chris Jones, U.S. Bank Corporate Treasury

Title: Trends in Mortgage Rate Modeling and the Prepayment Incentive

Abstract: Valuing mortgages in the context of the secondary market typically refers valuation of a mortgage-backed security (MBS) or mortgage servicing rights (MSR) asset. One fundamental challenge of this exercise is that model development relies on capturing market processes rooted in behavioral dynamics. In this talk, we will briefly review how MBS and MSRs are constructed, important modeling components, and then discuss how recent trends pre- and post-COVID have impacted models for these instruments.

Bio: Chris Jones is a model development manager at U.S. Bank focusing on the secondary mortgage market. He has 10 years of experience in financial modeling ranging from model risk management, balance sheet modeling, and the corporate investment portfolio. He holds a PhD in mathematics from the University of Pittsburgh.

Zoom Recording

Date: March 18, 2022

Speaker:  David H Bailey, Lawrence Berkeley National Lab (retired) and University of California, Davis

Title:  Why financial research is prone to false statistical discoveries

Abstract:  It is a sad fact that few investment funds, models or strategies actually beat the overall market averages over, say, a 10-year window.  Even in academic research work, care must be taken to avoid statistical pitfalls, because:  (a) the chances of finding a truly profitable investment design or strategy is very low, due to intense competition;  (b) true findings are mostly short-lived, as a result of the non-stationary nature of most financial systems;  and (c) it is often difficult to debunk a false claim.  Backtest overfitting is a particularly acute problem in finance, both in academic research and commercial development, since it is a simple matter to use a computer program to search thousands, millions or even billions of parameter or weighting variations to find an “optimal” setting.  In this talk, we summarize many of these pitfalls, explore why they are so prevalent, and present some tools that can be used to avoid them, including the “False strategy theorem”.

Bio: David H. Bailey (recently retired from the Lawrence Berkeley National Laboratory and also with the University of California, Davis) has published studies in computational mathematics, high-performance computing and mathematical finance.  He has received the Chauvenet and Merten Hesse Prizes from the Mathematical Association of America, the Levi Conant Prize from the American Mathematical Society, the Sidney Fernbach Award from the IEEE Computer Society and the Gordon Bell Prize from the Association for Computing Machinery.  He and his colleague Marcos López de Prado (Cornell University and Abu Dhabi Investment Authority) have published several studies highlighting the dangers of backtest overfitting and other statistical difficulties in mathematical finance.  Bailey is editor of the Mathematical Investor blog (https://www.mathinvestor.org)

Co-author:  Marcos Lopez de Prado, Cornell University and Abu Dhabi Investment Authority

Viewgraph file:  https://www.davidhbailey.com/dhbtalks/dhb-risk-2022.pdf

Zoom Recording

Date: March 4, 2022

Speaker: Turan G. Bali

Title: Machine Forecast Disagreement and Equity Returns

Abstract: We propose a novel measure of divergence of opinion among investors about stock value based on the dispersion in machines’ expected return forecasts. Compared to financial analysts, machines have neither behavioral biases nor conflicts of interest, thus we argue that machine forecast disagreement provides an objective measure of investor disagreement. After introducing a new measure of firm-specific uncertainty proxied by the degree of disagreement of machines’ future return forecasts, we show that this newly proposed, objective measure of uncertainty (or investor disagreement) does have a significant impact on the cross-sectional pricing of individual stocks. We find a significantly negative cross-sectional relation between machine forecast disagreement (MFD) and future stock returns. A long-short portfolio of stocks sorted by MFD provides a six-factor Fama-French (2018) alpha of 0.72% (1.06%) per month for the value-weighted (equal-weighted) portfolio. The return predictability is driven by mispricing rather than compensation for risk. The disagreement premium is also stronger for stocks that are largely held by retail investors, that receive less investor attention and that are costlier to arbitrage.

Bio: Turan G. Bali is the Robert Parker Chair Professor of Finance at the McDonough School of Business at Georgetown University. He received his Ph.D. from the Graduate School and University Center of the City University of New York in 1999. He also held visiting faculty positions at New York University and Princeton University. Professor Bali specializes in asset pricing, risk management, fixed income securities, and financial derivatives. A founding member of the Society for Financial Econometrics, he worked on consulting projects sponsored by major financial institutions and government organizations in the U.S. and other countries. He regularly presents his work at central banks, regulatory agencies, investment banks, hedge funds, and academic conferences. Professor Bali published three books and more than 50 articles in economics and finance journals, including the most prestigious journals in his field such as the Journal of Finance, Journal of Financial Economics, Review of Financial Studies, Journal of Monetary Economics, Management Science, Journal of Financial and Quantitative Analysis, Journal of Business, and Review of Economics and Statistics. He has won several awards, including the Q-Group's Jack Treynor Prize for quantitative research in finance. He currently serves as an Associate Editor of Management Science, Journal of Financial and Quantitative Analysis, Journal of Banking and Finance, Financial Management, and Journal of Portfolio Management. He also serves on the review committees of the National Science Foundation, Research Grants Council of Hong Kong, Social Sciences and Humanities Research Council of Canada, and Scientific and Technological Research Council of Turkey.

Date: February 25, 2022

Speaker: Lorenzo Torricelli

Title: From option values to additive models

Abstract: We have recently found that certain simple no-arbitrage vanilla option values yield to implied price distributions of logistic type which are known to be infinitely-divisible. When a no-arbitrage term function is also supplied, the corresponding family of distributions  determines an additive pure jump process for the underlying security price, which turns out to be a martingale. The use of additive processes in finance dates back to little more than a decade and has proved to successfully model derivative prices on a large number of asset classes. We insert in such literature with a focus on parameter parsimony and simplicity of valuation, while at the same time being able to capture returns  skewness, kurtosis,  self-similarity and other important stylized facts.  

Presentation

Zoom Recording

Date: February 18, 2022

Speaker: Elisa Alòs

Title: The asymptotic expansion of the regular discretization error of Itô integrals

Abstract: In this talk, we present an Edgeworth-type refinement of the central limit theorem for the discretization error of Itô integrals. Toward this end, we introduce a new approach, based on the anticipating Itô formula. This alternative technique allows us to compute explicitly the terms of the corresponding expansion formula. Two applications to finance are given: the asymptotic of discrete hedging error under the Black-Scholes model and the difference between continuously and discretely monitored variance swap payoffs under stochastic volatility models. Both of these applications shed light on hedging errors usually neglected in the continuous-time framework of mathematical finance. A short introduction to the anticipating Itô's calculus is given at the beginning of the talk. (Joint work with Masaaki Fukasawa).

Bio: Elisa Alòs is an associate professor in the Department of Economics and Business at Universitat Pompeu Fabra (UPF) and a BSE Affiliated Professor. Prior to joining the UPF in January 2001, she was an assistant professor at Universitat Autònoma de Barcelona (UAB). She completed her Ph.D. in Mathematics in 1998 at the University of Barcelona with a dissertation based on Malliavin Calculus techniques applied to the study of stochastic integral equations.

Her research relies on the applications of stochastic analysis in mathematical finance. In particular, it is focused on the application of Malliavin calculus techniques and the use of fractional noises in market modeling. Her main published results are related to the construction of closed-form approximation formulas for option prices for vanilla and exotic options, as well as with the analytical study of the properties of models (for example, the analytical study of the implied volatility skew for stochastic volatility models). Alòs is also associate editor of the SIAM Journal on Financial Mathematics.

Zoom Recording

Date: February 4, 2022

Speaker: Kateryna Holland

Title: Dynamics of Arbitrage

Abstract: We study the dynamics of cash-and-carry arbitrage using the U.S. crude oil market. Sizable arbitrage-related inventory movements occur at the New York Mercantile Exchange (NYMEX) futures contract delivery point but not at other storage locations, where instead, operational factors explain most inventory changes. We add to the theory-of-storage literature by introducing two new features. First, due to arbitrageurs contracting ahead, inventories respond to not only contemporaneous but also lagged futures spreads. Second, storage capacity limits can impede cash-and-carry arbitrage, leading to the persistence of unexploited arbitrage opportunities. Our findings suggest that arbitrage-induced inventory movements are, on average, price stabilizing.

Bio: Dr. Holland’s research interests are in the area of corporate finance and focus primarily on ownership, government involvement with firms, university innovation, various event studies, and energy. She has published in leading peer-reviewed journals such as the Journal of Financial Economics and the Journal of Corporate Finance. Professor Holland has earned numerous teaching awards and has taught corporate finance at the undergraduate, Masters and PhD levels.  She has worked as a power trader prior to earning her Ph.D. degree in Finance 

Date: January 28, 2022

Speaker: Teemu Pennanen

Title: Economic valuation of defined benefit pension liabilities

Abstract: The pensions industry operates largely by pricing annuities and other pension products with the actuarial discounting principle which can be traced back to at least the 19th century. This is in sharp contrast with financial economics and the banking industry where valuations are based on the costs of producing a product's payouts in the face of uncertainties and incompleteness of financial markets. The deficiencies of the classical actuarial methods are widely recognized by the industry but the transformation of the practices has been slow due to the challenges of applying economic valuation principles in pensions. Typical pension liabilities extend over several decades and their payouts depend on longevity developments which are uncertain and largely independent of the investment returns that insurers earn on their funds in financial markets. 

We present mathematical models and computational techniques for asset-liability management and valuation of defined benefit liabilities. The valuations look for the cheapest hedging strategy that covers the pension payments until maturity with an acceptable level of risk. Under complete markets assumption, this coincides with the classical replication argument while in the deterministic case, we recover the actuarial “best estimate”. The approach is illustrated by the valuation of the insurance portfolio of the Finnish private sector pension system.

Bio: Teemu Pennanen is a Professor of Financial Mathematics, probability and statistics at King's College London. Before joining KCL, professor Pennanen worked as Managing Director at QSA Quantitative Solvency Analysts Ltd, with a joint appointment as Professor of Mathematics at the University of Jyvaskyla. His research interests include convex optimization, probability and statistics and their applications to financial economics and risk management. Pennanen has authored over 50 journal publications and he has been a consultant to a number of financial institutions including Bank of Finland, The State Pension Fund and Ministry of Social Affairs and Health.

Presentation Link

Date: December 10, 2021

Speaker: Runhuan Feng

Title: A Unified Theory of Decentralized Insurance

Abstract: Decentralized insurance can be used to describe risk sharing mechanisms under which participants trade risks among each other as opposed to passing risks mostly to an insurer in traditional centralized insurance. There are a wide range of decentralized practices in all kinds of forms developed around the world, including online mutual aid in East Asia, takaful in the Middle East, peer-to-peer insurance in the West, international catastrophe risk pooling, etc. There is also a rich literature of risk sharing in academia that offers other decentralized mechanisms. This work presents a unified mathematical framework to describe the commonalities and the relationships of all these seemingly different business and theoretical models. Such a framework provides a fertile ground for the design and the analysis of hybrid and innovative models.

Bio: Runhuan Feng is a Professor of Mathematics, Statistics, Industrial and Enterprise Systems Engineering, Director of Actuarial Science Program, Director of Predictive Analytics & Risk Management Program, the State Farm Companies Foundation Professorial Scholar at the University of Illinois at Urbana-Champaign. He is the Faculty Lead for Finance and Insurance Sector at the University of Illinois System’s Discovery Partner Institute in Chicago. Runhuan is a Fellow of the Society of Actuaries and a Chartered Enterprise Risk Analyst. As an applied scientist, Runhuan strongly believes that most interesting research problems are discovered in response to the changing needs of the industry and the society. Runhuan’s research has been recognized in the practitioners' community through his applied technical contributions and presentations as invited speakers at industry conferences. His consulting work has been used by the Illinois General Assembly for pension-related legislative proposals.

Date: December 3, 2021

Speaker: Austin Pollok

Title: Kelly Criterion: From a Simple Random Walk to Levy Processes

Abstract: The original Kelly criterion provides a strategy to maximize the long-term growth of winnings in a sequence of simple Bernoulli bets with an edge, that is, when the expected return on each bet is positive. The objective of this work is to consider more general models of returns and the continuous time, or high-frequency, limits of those models. The results include an explicit expression for the optimal strategy in several models with continuous time compounding. Given we know how to optimally bet, we seek to find an edge in the financial markets by investigating the volatility risk premium in option returns. With the aid of high frequency volatility forecasts, we are able to capture an economically significant increase in risk premium compared to competing models.

Bio: Austin Pollok is a PhD student in Applied Mathematics at USC, set to graduate this year. His areas of research are in optimal growth strategies, such as the Kelly Criterion, under heavy-tailed processes, high frequency volatility forecasting using machine learning methods, as well as empirical option pricing. He has worked at Capital Group Companies as a quantitative research engineer while completing his PhD.

November 19, 2021

Speaker: Joseph Ivivi Mwaniki

Abstract: Popular models of finance, fall short of  accounting  for  most empirically  found stylized features of financial time series data, such as volatility clustering, skewness  and leptokurtic nature of log returns. In this study we propose a general framework for modeling  asset  returns which  account for serial dependencies in higher  moments and  leptokurtic nature  of  scaled GARCH filtered residuals.  Such  residuals are calibrated to normal inverse Gaussian and hyperbolic  distribution. Dynamics of risky assets assumed  in  Black Scholes model, Duans(1995) GARCH model  and other benchmark models for option valuation, are shown to be  nested in the proposed framework. Different sets of data are  used to support  the proposed framework.

Bio: https://profiles.uonbi.ac.ke/jimwaniki/content/dr-mwaniki-joseph-ivivi

November 12, 2021

Speaker: Yao Deng

Abstract: The standard investment-based asset pricing model predicts that stock returns equal investment returns, state-by-state. Yet, typical work in asset pricing only tests the weaker prediction that stock returns and investment returns should be equal on average. We document that by following the traditional methodology of only matching mean moments to estimate the model, most of the time series variation of stock returns is captured by the error terms, not by the predicted investment returns. We then show how to incorporate the model-implied time series restrictions in the estimation and testing of the model using the generalized method of moments, and formulate an external validity specification test. Our method uncovers a tradeoff between cross sectional fit and time series fit: the baseline investment-based model cannot fit both sets of moments simultaneously. Simulation exercises show that our estimation approach improves the power of the standard tests to detect model misspecification, and hence can be useful for improving the specification of future investment-based asset pricing models.

Bio: Yao Deng is an Assistant Professor of Finance at the University of Connecticut. He earned his PhD in Finance and Master in Financial Mathematics from the University of Minnesota. His research interests include empirical and theoretical asset pricing, behavioral finance, and macro finance.

November 5, 2021

Speaker: Patrick Guge Oloo Weke

Abstract: Claims reserving for general insurance business has developed significantly over the recent past. There has always been a slight mystery in short-term insurance contracts of how to go about reserving for claims, which have not yet come in, and are still in some sense of figment of the future. Stochastic models for triangular data are derived and applied to claims reserving data. The standard actuarial technique, the chain ladder technique is given a sound statistical foundation and considered as a linear model. The chain ladder technique and the two-way analysis of variance are employed for purposes of estimating and predicting the IBNR claims reserves. Insurance claims variables are non-normally distributed and therefore a measure that will capture the dependence among the variables better than the usual correlation is employed. One such method is the use of copulas. 

Bio: Patrick Weke is a full Professor of Actuarial Science and Financial Mathematics at the School of Mathematics, University of Nairobi since 2014.  He graduated with a B.Sc. (Honours) in Mathematics, Statistics and Computer Science in 1986 from University of Nairobi, an M.Sc. (Mathematical Statistics) in 1988 from University of Nairobi, an M.Sc. (Actuarial Science) in 1992 from The City University, London and a Ph.D. (Applied Actuarial Statistics) in 2001 from Harbin Institute of Technology, China. He is an Honorary Fellow of The Actuarial Society of Kenya (since 2017) and he is also involved in the following academic/professional activities: 

• Director, School of Mathematics, University of Nairobi (2014 – 2020) 
• Head, Actuarial Science and Financial Mathematics Division (2006 – 2014)
• Advisory Committee Member – Barclays Africa Chair in Actuarial Science, University of Pretoria, South Africa (2013 to date). 
• Director, UAP Life Assurance Ltd

He has successfully supervised 11 PhD candidates and over 40 MSc candidates in Actuarial Science and Financial Mathematics. He has published over 50 articles in refereed journals, 2 textbooks, 3 chapters in books and 20 conference proceedings. 

Speaker: Guojun Gan

Abstract: The healthcare sector in the U.S. is complex and is also a large sector that generates about 20\% of the country's gross domestic product. Healthcare analytics has been used by researchers and practitioners to better understand the industry. In this talk, I will present our recent work about the use of Beta regression models to understand the variability of brand name drug utilization across different areas with the U.S. The models are fitted to public datasets obtained from the Medicare & Medicaid Services and  the Internal Revenue Service. Integrated Nested Laplace Approximation (INLA) is used to perform the inference. Some numerical results showing the performance of Beta regression models will also be presented.

Bio: Guojun Gan is an Associate Professor in the Department of Mathematics at the University of Connecticut, where he has been since August 2014. Prior to that, he worked at a large life insurance company in Toronto, Canada for six years and a hedge fund in Oakville, Canada for one year. He received a BS degree from Jilin University, Changchun, China, in 2001 and MS and PhD degrees from York University, Toronto, Canada, in 2003 and 2007, respectively. He is also a Fellow of the Society of Actuaries (FSA).  His research interests are in the interdisciplinary areas of actuarial science and data science.  

Presentation Link

October 1, 2021

Speaker: Catherine Donnelly

Title: Pooling longevity for a better retirement income: how many people are needed?

Abstract: Pooled annuity funds are a way of converting retirement lump sum into an income stream for life.   Their objective is to provide a stable lifetime income to their participants.  They rely on the pooling of the participants' longevity risk to do this.  The participants bear all of the longevity risk in a pooled annuity fund, rather than it being transferred to an insurance company.

In the talk, I start with why pooled annuity funds should be a decumulation option for retirement.  Then I discuss the recent results which Thomas Bernhardt (U. Manchester, UK) and I have produced on the number of people needed for a fund to deliver on its objective.

Bio: Catherine Donnelly is a professor in actuarial math at Heriot-Watt University, Edinburgh and Director of the Risk Insight Lab. An actuary who has worked in pension consultancies before entering academia, she has a keen interest in developing workable solutions to improve people’s financial situation in retirement.  She has a PhD in financial mathematics from University of Waterloo, an MSc from University of Oxford and an undergraduate degree in mathematics from University of Cambridge.

Presentation Link

September 24, 2021

Speaker: Frederic Godin

Title: A mixed bond and equity fund model for the valuation of segregated fund policies

Abstract: Segregated fund and variable annuity policies are typically issued on mutual funds invested in both fixed income and equity asset classes. However, due to the lack of specialized models to represent the dynamics of fixed income fund returns, the literature has primarily focused on studying long-term investment guarantees on single-asset equity funds. This article develops a mixed bond and equity fund model in which the fund return is linked to movements of the yield curve. Theoretical motivation for our proposed specification is provided through an analogy with a portfolio of rolling horizon bonds. Moreover, basis risk between the portfolio return and its risk drivers is naturally incorporated into our framework. Numerical results show that the fit of our model to segregated fund data is adequate. Finally, the valuation of segregated fund policies is illustrated and it is found that the interest rate environment can have a substantial impact on guarantee costs.

Bio: I am an Associate Professor at the Mathematics and Statistics Department of Concordia University in Montreal, Quebec, Canada. My research interests are financial engineering, risk management, actuarial science, reinforcement learning, stochastic modeling, dynamics programing, variable annuities and energy markets. I hold the Fellow of the Society of Actuaries (FSA) and Fellow of the Canadian Institute of Actuaries (FCIA) designations. I am part of the Quantact research group.

Presentation Link

April 16, 2021

Speaker: Dilip Madan

Title: Efficient Exposure Frontiers

Abstract: Risk is described by the instantaneous exposure to changes in valuations induced by the arrival rate of economic shocks. The arrival rate mea- sure is typically not a probability measure and often the aggregate arrival rate across all shocks is infinite. Risk management and portfolio theory are conse- quently recast as managing this exposure risk. There is no risk free exposure with all fixed income securities subject to the risks of instantaneous changes in their valuations. The reference return in the economy is that of a zero risk gra- dient return, typically estimated as negative. Required returns on assets with low risk gradients are then negative. It is also observed that required returns are robust to positions on the efficient frontier as well the construction of the frontier itself. Both equity and fixed income security frontiers are constructed as illustrations of efficient risk positions.

BIO: Dilip Madan is Professor of Finance at the Robert H. Smith School of Business. He specializes in Mathematical Finance. Currently he serves as a consultant to Morgan Stanley, Meru Capital and Caspian Capital. He has also consulted with Citigroup, Bloomberg, the FDIC and Wachovia Securities. He is a founding member and Past President of the Bachelier Finance Society. He received the 2006 von Humboldt award in applied mathematics, was the 2007 Risk Magazine Quant of the year, received the 2008 Medal for Science from the University of Bologna and held the 2010 Eurandom Chair. He is Managing Editor of Mathematical Finance, Co-editor of the Review of Derivatives Research, Associate Editor of the Journal of Credit Risk and Quantitative Finance. His work is dedicated to improving the quality of financial valuation models, enhancing the performance of investment strategies, and advancing the efficiency of risk allocation in modern economies. Recent major contributions have appeared inMathematical Finance, Finance and Stochastics, Quantitative Finance, the Journal of Computational Finance, The International Journal of Theoretical and Applied Finance, The Journal of Risk, The Journal of Credit Risk among other journals.

Presentation Link

March 26, 2021

Speaker: Jean-François Bégin

Title: Do Jumps Matter in the Long Term? A Tale of Two Horizons

Abstract: Economic scenario generators (ESGs) for equities are important components of the valuation and risk management process of life insurance and pension plans. Because the resulting liabilities are very long-lived and the short-term performance of the assets backing these liabilities may trigger important losses, it is thus a desired feature of an ESG to replicate equity returns over such horizons. In light of this horizon duality, we investigate the relevance of jumps in ESGs to replicate dynamics over different horizons and compare their performance to popular models in actuarial science. We show that jump-diffusion models cannot replicate higher moments if estimated with the maximum likelihood. Using a generalized method of moments-based approach, however, we find that simple jump-diffusion models have an excellent fit overall (moments and the entire distribution) at different time scales. We also investigate three typical applications: the value of one dollar accumulated with no intermediate monitoring, a solvency analysis with frequent monitoring, and a dynamic portfolio problem. We find that jumps have long-lasting effects that are difficult to replicate otherwise, so yes, jumps do matter in the long term.

This is joint work with Mathieu Boudreault.

Bio: Jean-François Bégin, PhD, FSA, FCIA is an Assistant Professor in the Department of Statistics and Actuarial Science at Simon Fraser University. His research interests include financial modelling, financial econometrics, filtering methods, high-frequency data, credit risk, option pricing, and pension economics. Before joining SFU, he received his PhD from HEC Montréal.

Presentation Link

March 19, 2021

Speaker: Liang (Jason) Hong

Title: Model misspecification, Bayesian versus credibility estimation, and Gibbs posteriors

Abstract: In the context of predicting future claims, a fully Bayesian analysis – one that specifies a statistical model, prior distribution, and updates using Bayes’s formula – is often viewed as the gold-standard, while Bühlmann’s credibility estimator serves as a simple approximation. But those desirable properties that give the Bayesian solution its elevated status depend critically on the posited model being correctly specified. Here we investigate the asymptotic behavior of Bayesian posterior distributions under a misspecified model, and our conclusion is that misspecification bias generally has damaging effects that can lead to inaccurate inference and prediction. The credibility estimator, on the other hand, is not sensitive at all to model misspecification, giving it an advantage over the Bayesian solution in those practically relevant cases where the model is uncertain. This begs the question: does robustness to model misspecification require that we abandon uncertainty quantification based on a posterior distribution? Our answer to this question is No, and we offer an alternative Gibbs posterior construction. Furthermore, we argue that this Gibbs perspective provides a new characterization of Bühlmann’s credibility estimator.

Bio: Liang Hong, PhD, FSA, is an Associate Professor in the Department of Mathematical Sciences at the University of Texas at Dallas. His current research interests are actuarial science and foundations of mathematics. In actuarial science, he is primarily interested in applying machine/statistical learning methods, such as Bayesian non-parametric models, conformal prediction, and Gibbs posteriors, to solve important insurance problems.

Presentation Link

March 12, 2021

Speaker: Mathieu Rosenbaum 

Title: Rough Volatility

Abstract: The goal of this talk is to introduce rough volatility models. We will demonstrate that this approach significantly outperforms conventional ones, both from a statistical and a risk management viewpoint. We will notably illustrate this showing how this new class of models enables us to solve long standing problems in financial engineering.

Bio: Mathieu Rosenbaum is a full professor at École Polytechnique, where he holds the chair “Analytics and Models for Regulation” and is co-head of the quantitative finance (El Karoui) master program. His research mainly focuses on statistical finance problems, regulatory issues and risk management of derivatives. He published more than 65 articles on these subjects in the best international journals. He is notably one of the most renowned experts on the quantitative analysis of market microstructure and high frequency trading.  On this topic, he co-organizes every two years in Paris the conference "Market Microstructure, Confronting Many Viewpoints".  He is also at the origin (with Jim Gatheral and Thibault Jaisson) of the development of rough volatility models. Mathieu Rosenbaum has collaborations with various financial institutions (investment banks, hedge funds, regulators, exchanges...), notably BNP-Paribas since 2004. He also has several editorial activities as he is one of the editors in chief of the journal “Market Microstructure and Liquidity“ and is associate editor for 10 other journals.  He received the Europlace Award for Best Young Researcher in Finance in 2014, the European Research Council Grant in 2016, the Louis Bachelier prize in 2020 and the Quant of the Year award in 2021.

Presentation Link

March 5, 2021

Speaker: Xiaobai Zhu

Title: Cyclical Design for Target Benefit Pension Plan

Abstract: In this paper, we derived the optimal cyclical design of Target Benefit (TB) pension plan. We focused on the stability of the benefit payment, and formulated an optimal control problem using a regime-switching model. We drew a number of remarks to improve the readability of our explicit solution, and made simplifications to enhance the transparency of the risk sharing design. We provided a new yet natural interpretation for a commonly used parameter under the TB context. We highlighted that cautions must be made when studying TB design using optimal control theory. Our numerical result suggested that a 100/0 investment strategies is preferred for the robustness of TB design, and the risk sharing mechanism should include both counter- and pro-cyclical components.

Bio: For my personal information, my full name is Xiaobai Zhu, I am assistant professor at School of Insurance, Southwestern University of Finance and Economics, China, my research interest is on hybrid pension plans and longevity modelling.

Presentation Link

February 26, 2021 * 9 am CDT -  Seminar Canceled - To Be Rescheduled at a Later Date

Speaker: Justin Sirignano

Title: Deep Learning Models of High-Frequency Financial Data

Abstract: We develop and evaluate deep learning models for predicting price movements in high-frequency data. Deep recurrent networks are trained on a large limit order book dataset from hundreds of stocks across multiple years. Several data augmentation methods to reduce overfitting are analyzed. We also develop and evaluate deep reinforcement learning models for optimal execution problems with limit order book data. "Optimal execution" is the problem of formulating, given an a priori determined order direction (buy or sell) and order size, the optimal adaptive submission strategy to complete the order at the best possible price(s).The performance of deep recurrent models is compared against other types of models trained with reinforcement learning, such as linear VAR models and feedforward neural networks.

Bio: Justin Sirignano is an Associate Professor at the Mathematical Institute at the University of Oxford, where he is a member of the Mathematical & Computational Finance and Data Science groups. He received his PhD from Stanford University and was a Chapman Fellow at the Department of Mathematics at Imperial College London. His research interests are in the areas of applied mathematics, machine learning, and computational methods.

February 19, 2021

Speaker: Sergei Levendorskii 

Title: Static and semi-static hedging as contrarian or conformist bets

Abstract: Once the costs of maintaining the hedging portfolio are properly takeninto account, semi-static portfolios should more properly be thought of as separate classes of derivatives, with non-trivial, model-dependent payoff structures. We derive new integral representations for payoffs of exotic European options in terms of payoffs of vanillas, different from the Carr-Madan representation, and suggest approximations of the idealized static hedging/replicating portfolio using vanillas available in the market. We study the dependence of the hedging error on a model used for pricing and show that the variance of the hedging errors of static hedging portfolios can be sizably larger than the errors of variance-minimizing portfolios. We explain why the exact semi-static hedging of barrier options is impossible for processes with
jumps, and derive general formulas for variance-minimizing semi-static portfolios. We show that hedging using vanillas only leads to larger errors than hedging using vanillas and first touch digitals. In all cases, efficient calculations of the weights of the hedging portfolios are in the dual space using new efficient numerical methods for calculation of the Wiener-Hopf factors and Laplace-Fourier inversion.

Bio: Dr. Levendorskii is a founding partner at Calico Science Consulting in Austin TX. Dr. Levendorskii has developed several models and methods used by the financial services industry. His areas of expertise are Lévy processes with heavy and semi-heavy tails, Financial Mathematics, Real Options, Stochastic Optimization, Applied Fourier Analysis, Spectral Theory, Degenerate Elliptic Equations, Pseudo-differential operators, Numerical methods, Insurance, Quantum Groups, and Fractional Differential Equations. Prior to Calico, he was Chair in Financial Mathematics and Actuarial Sciences, Department of Mathematics and Deputy Director of Institute of Finance, University of Leicester, United Kingdom. He holds a Doctor of Sciences in Mathematics from Academy of Sciences of the Ukraine and he also earned a PhD in Mathematics from Rostov State University."

Presentation Link

February 12, 2021 

Speaker: Indranil SenGupta

Title: A machine learning-driven crude oil data analysis, with applications in continuous-time quadratic hedging

Abstract: In this presentation, a refined Barndorff-Nielsen and Shephard (BN-S) model is implemented to find an optimal hedging strategy for commodity markets. The refinement of the BN-S model is obtained through various machine and deep learning algorithms. The refinement leads to the extraction of a deterministic parameter from the empirical data set. The analysis is implemented to the Bakken crude oil data and the aforementioned deterministic parameter is obtained for a wide range of data sets. With the implementation of this parameter in the refined model, it is shown that the resulting model performs much better than the classical stochastic models.

Bio: Indranil SenGupta is an Associate Professor at the Department of Mathematics at North Dakota State University (NDSU). He is currently the mathematics graduate program director at NDSU. He received his Ph.D. in mathematics from Texas A&M University in 2010. His research interests include mathematical finance, stochastic processes, and data-science. He was the Associate Editor-in-Chief of the journal Mathematics, 2014-2019. Currently, he is an associate editor in the area of finance and risk management for the Journal of Modelling in Management. He is in the editorial board for several other journals.

Presentation Link

February 5, 2021 

Speaker: Sandra Paterlini

Title: Sorting out your investments: sparse portfolio selection via the sorted l1-norm

Abstract: We introduce a financial portfolio optimization framework that allows us to automatically select the relevant assets and estimate their weights by relying on a sorted l1-Norm penalization, henceforth SLOPE. To solve the optimization problem, we develop a new efficient algorithm, based on the Alternating Direction Method of Multipliers. SLOPE is able to group constituents with similar correlation properties, and with the same underlying risk factor exposures. Depending on the choice of the penalty sequence, our approach can span the entire set of optimal portfolios on the risk-diversification frontier, from minimum variance to the equally weighted. Our empirical analysis shows that SLOPE yields optimal portfolios with good out-of-sample risk and return performance properties, by reducing the overall turnover, through more stable asset weight estimates. Moreover, using the automatic grouping property of SLOPE, new portfolio strategies, such as sparse equally weighted portfolios, can be developed to exploit the data-driven detected similarities across assets.

Bio: Sandra Paterlini is full professor at the University of Trento, Italy. From 2013 to 2018, she held the Chair of Financial Econometrics and Asset Management at EBS Universität für Wirtschaft und Recht, Germany. Before joining EBS, she was assistant professor in statistics at the Faculty of Economics at the University of Modena and Reggio E., Italy. From 2008 to 2012, she has been a long-term visiting professor at the School of Mathematics, University of Minnesota. Her research on financial econometrics, statistics, operational research and machine learning have been predominantly interdisciplinary and often with an applied angle. Her work experience as a business consultant in finance and as a collaborator of central banks, such as for European Central Bank, Deutsche Bundesbank and the Fed Cleveland, has given her valuable input to guide and validate her research. Furthermore, she spent many years abroad (US, Germany, UK, and Denmark) to broaden and improve her skills further and to establish an international network of collaborators. She has been a consultant on business projects related to style analysis, portfolio optimization and risk management.

Her latest research interests are on machine learning methods for asset allocation, network analysis, risk management and ESG.

Presentation Link

January 29, 2021 

Speaker: Margie Rosenberg, University of Wisconsin - Madison

Title: A Cluster Analysis Application Using only Social Determinant Variables to Predict Profiles of US Adults having the Highest Health Expenditures

Abstract: Social determinants of health are defined as the social and physical conditions in which people are born, grow, live, work and age that impact health outcomes.  In the late 1960s, Andersen developed a behavioral health framework to help shape a discussion of the impact of social determinants on medical  services and other outcomes. Andersen and Newman acknowledged that some populations were not  receiving, nor having access to, the same level of medical care as other populations. Our work focuses on social determinants and examining their impact on health expenditures of working  aged US adults (20 – 59).  We use  longitudinal data that are nationally representative of the US adult  working‐age civilian non‐institutionalized population.  Our study includes Individuals who participated in  the National Health Interview Study (NHIS), and who are included in the following two years of the  Medical Expenditure Panel Study (MEPS). We form clusters based on the 2010 NHIS demographic, economic, and health‐related characteristics  that are commonly used in studies of health care utilization. We use data from the 2010 NHIS cohort to  create clusters using a clustering algorithm called Partitioning Around Medoids. Health expenditure  distributions for this cohort are examined over the following two years. We validate the approach by applying the centers of the clusters to the 2008 and 2009 NHIS cohorts. Finally, we examine the  effectiveness of these clusters in representing the top 5% of health care utilizers. Our findings show that these clusters can provide health care organizations a sampling approach to  perform a first‐stage audit using a small segment of the population that can help identify the highest of the utilizers. The approach also identifies those who do not have health expenditures that could signal  underutilization. While the profiles designed are representative of US adults, the approach can be applied to any population to reveal the impact of the profiles on utilization. Clusters formed using the data without comorbidities can profile new insureds to allow prospective management of certain  individuals. The same group profiles can be used in multiple studies with different outcomes, such as  inpatient, outpatient, or drug expenditures.

Bio: Margie Rosenberg, PhD, FSA is the Assurant Health Professor of Actuarial Science Professor at the University of Wisconsin-Madison. Margie’s research interests are in the application of statistical methods to health care, and applying her actuarial expertise to cost and policy issues in health care. Her recent research involves linking social determinants to outcomes such as (i) assessing the impact of delayed attention to oral health issues on emergency department visits and (ii) assessing the impact of unhealthy behaviors on perceived health status and predicting individuals with persistent high expenditures. Prior to her starting on her academic career, Margie worked as a life actuary for Allstate Life Insurance Company in Northbrook, IL.

Presentation Link

November 20, 2020

Speaker: David Matteson, Affiliation: Cornell

Title: Dynamic Shrinkage Processes

Abstract: We propose a novel class of dynamic shrinkage processes for Bayesian time series and regression analysis. Building on a global–local framework of prior construction, in which continuous scale mixtures of Gaussian distributions are employed for both desirable shrinkage properties and computational tractability, we model dependence between the local scale parameters. The resulting processes inherit the desirable shrinkage behaviour of popular global–local priors, such as the horseshoe prior, but provide additional localized adaptivity, which is important for modelling time series data or regression functions with local features. We construct a computationally efficient Gibbs sampling algorithm based on a Pólya–gamma scale mixture representation of the process proposed. Using dynamic shrinkage processes, we develop a Bayesian trend filtering model that produces more accurate estimates and tighter posterior credible intervals than do competing methods, and we apply the model for irregular curve fitting of minute‐by‐minute Twitter central processor unit usage data. In addition, we develop an adaptive time varying parameter regression model to assess the efficacy of the Fama–French five‐factor asset pricing model with momentum added as a sixth factor. Our dynamic analysis of manufacturing and healthcare industry data shows that, with the exception of the market risk, no other risk factors are significant except for brief periods. If time permits, we will also highlight extensions to change point analysis and adaptive outlier detection.

Youtube link to Presentation
https://youtu.be/W62PbdMhz40

October 30, 2020 

Speaker: John Dodson, Options Clearing Corporation

Title: Trends in applied mathematics and its adoption in the finance industry, or why you should pass on blockchains and big data 

Abstract: Over the course of the twentieth century, applied mathematics has gradually assimilated and standardized the subjects of probability, statistics, control, and information. While an outside observer of decadal trends in STEM in finance might instead focus on the industry's embrace of computing technology during the Moore's Law era, I claim these quieter developments are ultimately more impactful because they help firms to organize information technology and financial innovation to create lasting value for clients. I will demonstrate this through a survey of the changing role of quants, and make an attempt to describe current opportunities.

October 23, 2020 

Speaker: Shae Armstrong, Optum

Title: Quantifying the Impact of the Social Determinants of Health in the Covid-19 Era

Abstract: The Social Determinants of Health (SDoH) are key factors in each person’s environment and life that influence clinical outcomes of their health and wellbeing. These factors include, but are not limited to, income, housing, food security, education, and geography. In the age of Covid-19, understanding these factors and how they correlate to each other is more important than ever. Once we as industry gain insight on these clinical and financial impacts, we need to translate that insight into policy to mitigate root cause issues to better serve patients across the country.

During this lecture we lay the foundation by defining what the Social Determinants of Health are and the various categories they fall into. We will also examine what data sources feed various SDoH models and limitations of said data sources. Next we will conduct a deep-dive examination on a variety of case studies and models aimed at quantifying the short-term and long-term clinical and financial impact of Covid-19. From there we will touch on the future and impact of healthcare data analytics within the healthcare industry and as human beings navigating an unprecedented pandemic.

October 15, 2020

Speaker: Zhiguang (Gerald) Wang, South Dakota State University 

Title: Multi-Step Forecast of Implied Volatility Surface using Deep Learning

Abstract: Modeling implied volatility surface (IVS) is of paramount importance to price and hedge an option. We contribute to the literature by modeling the entire IVS using recurrent neural network architectures, namely Convolutional Long Short Term Memory Neural Network (ConvLSTM) to produce multivariate and multi-step forecasts of the S&P 500 implied volatility surface. Using the daily S&P 500 index options from 2002 to 2019, we benchmark the ConvLSTM model against traditional multivariate time series VAR model, VEC model, and LSTM neural network. We find that both LSTM and ConvLSTM can fit the training data extremely well with mean absolute percentage error (MAPE) being 3.56%  and 3.88%, respectively. As for out-of-sample data, the ConvLSTM (8.26% ) model significantly outperforms traditional time series models as well as the LSTM model for a 1-day, 30-day, and 90-day horizon, for all moneyness groups and contract months of both calls and puts.  

October 2, 2020

Speaker: Liban Mohammed, University of Wisconsin -Madison

Title: Efficient Risk-sensitivity Estimation for Equity-Linked Insurance Benefits

Abstract: For an organization with billions of dollars in assets, precise risk management is necessary to safeguard those assets. However, when the risks these assets are exposed to depend on the future performance of equities in complex ways, directly estimating them in real-time to the necessary precision can be prohibitively expensive. This talk discusses some approaches to resolving this tension via metamodeling techniques.

YouTube Link to Presentation
https://youtu.be/oQmX943Wmtg

Friday, September 25, 2020 

Speaker: Max Rudolph, Rudolph Financial

Title: Actuarial Implications of COVID-19

Abstract: COVID-19 has had a material impact on all practice areas of the actuarial profession, ranging widely include traditional areas like health and mortality claims, assets and economic activity, but also risk management and strategic planning. This session assumes you know many of the basic statistics and provides observations about how analysis of the virus is evolving. Bio: MAX J. RUDOLPH, FSA CFA CERA MAAA. Max Rudolph is a credentialed actuary, active in the Asset-Liability Management and Enterprise Risk Management space for many years. He was named a thought leader in ERM within the actuarial profession, chaired the ERM Symposium, the SOA Investment Section Council and the SOA’s Investment Actuary Symposium. He is a past SOA board member and received a Presidential Award for his role developing the CERA credential. He was the subject matter expert for the original Investment and ERM modules, wrote the ERM courseware document and has been involved with the actuarial profession’s climate change and pandemic efforts. He is a frequent speaker at actuarial seminars and universities, and an award-winning author. For the past 14 years Max has led Rudolph Financial Consulting, LLC, an independent consulting practice, focusing its insurance practice on ERM and ALM consulting. He has completed projects relating to life, health, annuity, and casualty insurers. He is an adjunct professor for Creighton University’s Heider School of Business, where he focuses on ERM and investment topics.Max has completed a number of well received research reports covering topics such as emerging risks, low growth, low interest rates, investments, systemic risk and ERM. Other topics he has written about include pandemics, ALM and value investing. Many of his papers can be found at www.rudolph-financial.com. He comments on a variety of risk topics from @maxrudolph on twitter.

YouTube Link to Presentation
https://youtu.be/Y32WtsaqL0E  

February 7, 2020 

Speaker: Kaisa Taipale, C.H. Robinson

Title: Pricing in Contractual Freight Compared to Finance

Abstract: In this talk, I’ll discuss the contractual freight business, in which a large shipper makes a contract with a company like CH Robinson to procure carriers (trucks) for their goods over the course of a year for a given rate, as opposed to using the volatile “spot” or transactional market. Because these year-long contracts aren’t legally binding, some shippers treat them more like an American option on the underlying price of freight — but this has game-theoretic economic consequences for the shipper! Dr. Taipale, Data Scientist at C.H. Robinson will also talk about the data science and mathematical skills that are important for her job at C.H. Robinson