Kjersti Aas's home page

Kjersti Aas

Dr. Philos
Research Director
Department of Statistical Analysis and Machine Learning.
Norwegian Computing Center ,
Gaustadalléen 23,
P.O. Box 114 Blindern,
N-0314 Oslo,
Norway
Tel: (+47) 22 85 25 00
Fax: (+47) 22 69 76 60
Kjersti.Aas@nr.no


Positions:

  • Research director in the department of Statistical Analysis and Machine Learning at the Norwegian Computing Center and
  • Head of research and marketing for the area : Statistical Analysis and Machine Learning for the finance, insurance and commodity markets.
  • Adjunct professor at the Norwegian University of Science and Technology from January 2021 to January 2024.
  • Adjunct professor at the University of Bergen from February 2011 to February 2017.

    Participitation in boards:

  • Member of the Board of Directors of Trondheim Kommunale Pensjonskasse from January 2016.
  • Member of the Board of Directors of MP pensjon from January 2008 to June 2013.

    Press (In Norwegian):

  • Roboten gir lån hvis den får gode vibrasjoner fra kontoen din , Dagens Næringsliv, 1. mai, 2017.
  • Gode modeller gir innsikt , Dagens Næringsliv, 27. juni, 2016.
  • Risikostyringen for boliglån er ikke god nok , Finansavisen, 6. juni, 2013.
  • Hvem sier at Basel I hadde sannheten? Finansavisen, 2. november, 2012.
  • Hvorfor så forskjellig i bank og forsikring? Finansavisen, 22. desember, 2011.
  • Ekstremvær, Bolt og børsfall , Dagens Næringsliv, 21. september, 2011.
  • Stor formue - flaks eller dyktighet , Kapital 18/10, 22. oktober 2010.
  • Intervju i boken TALENTER PÅ SPILL: Eksempler på god forskningsledelse ,utgitt av Komité for integreringstiltak - kvinner i forskning 07-10.
  • Intervju i Internrevisoren (medlemsbladet til Norges Interne Revisorers Forening), nr 1, 2009 (publisert 14. juli 2009).
  • Veldig lettvint av Oljefondet Intervju i Dagens Næringsliv 16. mars, 2009.
  • Risiko og krise Kommentar i Dagens Næringsliv 16. mars, 2009.
  • Ingen fri lunsj mer? Kommentar i Dagens Næringsliv 9. mars, 2009.
  • Hvor mange sorte svaner? Kommentar i Dagens Næringsliv 13. desember, 2008.
  • Historie og framtid kredittkommentar i Dagens Næringsliv 7. april, 2008.
  • Tvangssalg i aksjemarkedet, kredittkommentar i Dagens Næringsliv 11. februar, 2008.
  • Nye metoder - nye svar , kredittkommentar i Dagens Næringsliv 26. november, 2007.
  • Krakket som ikke kunne skje , intervju i Dagens Næringsliv 19. oktober, 2007.
  • Jernkontroll på risikoen, intervju i Byens Næringsliv 19/09/07.
  • Slik takler du risikofellene, intervju i Ukeavisen Ledelse 07/09/07.
  • Bankene bedre rustet mot tap, intervju i Ukeavisen Ledelse 28/08/07.
  • Bankene bedre rustet mot tap, intervju i Økonomisk Rapport 28/08/07.
  • Bankene bedre rustet, innlegg på forskning.no 27. august, 2007.
  • "K-ordet", omtale i Kapital 05/2007, 26. mars 2007.
  • "Er krakket rett rundt hjørnet?", intervju i Økonomisk Rapport 03/07, 15. februar.2007.
  • Ser ikke rasfaren , intervju i Kapital 16/05, 22.09.2005.
  • Sannsynligheten for et nytt børskrakk er kanskje større enn man tror , innlegg på forskning.no 13. september, 2005.

    International journal papers 2004-:

  • Aas, K., Charpentier, A., Huang, F., and Richman, R.: Insurance analytics: prediction, explainability, and fairness , Annals of Actuarial Science, Volume 18, pp. 535-539, 2024.
  • Redelmeier, Annabelle, Jullum, Martin, Aas, Kjersti and Løland, Anders: MCCE: Monte Carlo sampling of valid and realistic counterfactual explanations for tabular data , Data Mining and Knowledge Discovery, published online March 2024.
  • Olsen, Lars, H., Glad, I., Aas, K. and Jullum, M.: A Comparative Study of Methods for Estimating Conditional Shapley Values and When to Use Them ,Data Mining and Knowledge Discovery, published online March 2024.
  • Mancisidor, Rogelio A., Kampffmeyer, Michael, Aas, Kjersti and Jenssen, Robert: Discriminative Multimodal Learning via Conditional Priors in Generative Models , Neural Networks, Volume 169, Pages 417--430, January 2024,
  • Olsen, Lars, H., Glad, I., Aas, K. and Jullum, M.: Using Shapley Values and Variational Autoencoders to Explain Predictive Models with Dependent Mixed Features, Jornal of Machine Learning Research (JMLR), Vol. 23, pp. 1--51, 2022.
  • Mancisidor, Rogelio A., Kampffmeyer, Michael, Aas, Kjersti and Jenssen, Robert: Generating Customer's Credit Behavior with Deep Generative Models, , Knowledge-based systems, Volume 245, June 2022.
  • Wahl, Jens Cristian, Aanes, Fredrik Lohne and Aas, Kjersti: Spatial modelling of risk premiums for water damage insurance , Scandinavian Actuarial Journal, Vol. 2022, pp. 216--233, 2022.
  • Aas, Kjersti, Nagler, Thomas, Jullum, Martin and Løland, Anders: Explaining predictive models using Shapley values and non-parametric vine copulas . Dependence Modeling, Vol. 9, pp. 62--81, 2021.
  • Jullum, Martin, Redelmeier, Annabelle and Aas, Kjersti: Efficient and simple prediction explanations with groupShapley: A practical perspective , 2nd Italian workshop on Explainable Artificial Intelligence, Italy, December 1-3, 2021.
  • Aas, Kjersti, Jullum, Martin and Løland, Anders: Explaining individual predictions when features are dependent: More accurate approximations to Shapley values , Artificial Intelligence, Vol. 298, September 2021.
  • Mancisidor, Rogelio A., Kampffmeyer, Michael, Aas, Kjersti and Jenssen, Robert: Learning Latent Representations of Bank Customers With The Variational Autoencoder, Expert Systems and Applications, Volume 164, February 2021.
  • Redelmeier, Annabelle, Jullum, Martin and Aas, Kjersti: Explaining predictive models with mixed features using Shapley values and conditional inference trees , In: Holzinger A., Kieseberg P., Tjoa A., Weippl E. (eds) Machine Learning and Knowledge Extraction. CD-MAKE 2020. Lecture Notes in Computer Science, vol 12279. Springer, Cham., 2020.
  • Mancisidor, Rogelio A., Kampffmeyer, Michael, Aas, Kjersti and Jenssen, Robert: Deep Generative Models for Reject Inference in Credit Scoring . Knowledge-based systems, Volume 196, May 2020.
  • Aas, Kjersti and Rognebakke, Hanne: The evolution of a mobile payment solution network , Network Science, Vol 7, pp. 422-437, September, 2019.
  • Kvamme, H., Sellereite, N., Aas, K. and Sjursen, S.: Predicting mortgage default using convolutional neural networks, Expert Systems with Applications, Volume 102, Pages 207-217, 2018.
  • Aas, Kjersti, Neef, Linda R., Williams L.,Raabe, Dag: Interest rate model comparisons for participating products under Solvency II , Scandinavian Actuarial Journal, Volume 2018, No. 3, pp. 203-224, February 2018.
  • Aas, Kjersti: Pair-copula constructions for financial applications: A review , Econometrics, Volume 4, Number 4, October 2016.
  • Hobæk Haff, Ingrid, Aas, Kjersti, Frigessi, Arnoldo and Graziani, Virginia L.: Structure learning in BBNs using regular vines , Computational Statistics & Data Analysis, Volume 101, Issue C, pp. 186-208, September 2016.
  • Low, Rand Kwong Yew, Faff, Robert and Aas, Kjersti: Enhancing mean-variance portfolio selection by modeling distributional asymmetries , Journal of Economics and Business, Volume 85, pp. 49-72, May-June 2016.
  • Aas, Kjersti and Puccetti, Giovanni: Bounds for total economic capital: the DNB case study, Extremes, Volume 17, Issue 4, pp 693-715, 2014.
  • Aas, Kjersti, Neef, Linda R., Raabe, Dag and Vårli, Ingborg D.: A simulation-based ALM model in practical use by a Norwegian Life Insurance company, In Modern Problem in Insurance Mathematics , Silvestrov, Dmitrii, Martin-Löf, Anders (Eds.), Springer 2014.
  • Günther, Clara-Cecilie, Tvete, Ingunn Fride, Aas, Kjersti and Borgan, Ørnulf: Predicting future claims among high risk policyholders using random effects, In Modern Problem in Insurance Mathematics , Silvestrov, Dmitrii, Martin-Löf, Anders (Eds.), Springer 2014.
  • Günther, Clara-Cecilie, Tvete, Ingunn Fride, Aas, Kjersti, Sandnes, Geir Inge and Borgan, Ørnulf: Modelling and predicting customer churn from an insurance company, Scandinavian Actuarial Journal, Volume 2014 (1), pp 58-71, 2014.
  • Brechmann, Eike C., Czado, Claudia, Aas, Kjersti: Truncated regular vines in high dimensions with application to financial data , Canadian Journal of Statistics, Vol. 40, pp 68-85, No. 1, 2012.
  • Martino, Sara, Aas, Kjersti, Lindqvist, Ola, Neef, Linda R. and Rue, Håvard: Estimating Stochastic Volatility Models Using Integrated Nested Laplace Approximations , European Journal of Finance, Volume 17, Issue 7, 2011.
  • Cooke, Roger, Joe Harry and Aas, Kjersti, "Vines Arise", In DEPENDENCE MODELING: Handbook on Vine Copulae , D. Kurowicka and Harry Joe (eds.), World Scientific Publishing Co., February 2011.
  • Aas, Kjersti and Berg, Daniel, "Modelling dependence between financial returns using pair-copula constructions", In DEPENDENCE MODELING: Handbook on Vine Copulae , D. Kurowicka and Harry Joe (eds.), World Scientific Publishing Co., February 2011.
  • Reitan, Trond and Aas, Kjersti: A New Robust Importance Sampling Method for Measuring VaR and ES Allocations for Credit Portfolios, Journal of Credit Risk, Vol 6, No. 4, p. 1-37, 2010/2011.
  • Hobæk Haff, Ingrid, Aas, Kjersti and Frigessi, Arnoldo: On the simplified pair-copula construction - simply useful or too simplistic? , Journal of Multivariate Analysis, 101(5), 1296-1310, 2010.
  • Berg, Daniel and Aas, Kjersti: Models for construction of multivariate dependence, European Journal of Finance, 15 (7/8),639-659, 2009.
  • Aas, Kjersti: Discussion of ``Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations,'' by H. Rue, S. Martino and N. Chopin. JRSS-B, 71, Part2, 2009.
  • Aas, Kjersti, Czado, Claudia, Frigessi, Arnoldo and Bakken, Henrik: Pair-copula constructions of multiple dependence , Insurance: Mathematics and Economics, Vol. 44, No. 2, 2009 (doi:10.1016/j.insmatheco.2007.02.001).
  • Aas, Kjersti, Dimakos, Xeni K. and Øksendal, Anders: Risk Capital Aggregation , Risk Management 9(2) , April 2007.
  • Aas, Kjersti and Hobæk Haff, Ingrid: The Generalised Hyperbolic Skew Student’s t-distribution , Journal of Financial Econometrics,4(2), March 2006.
  • Aas, Kjersti, Hobæk Haff, Ingrid and Dimakos Xeni K: Risk Estimation using the Multivariate Normal Inverse Gaussian Distribution, Journal of Risk, 8(2), Winter 2005/2006.
  • Dimakos, X. K. and Aas, K, Integrated risk modeling, Statistical modeling, Vol. 4,1-13, 2004.
  • Aas, K. and Kåresen, K., The Matrix, Energy Power Risk Management, 9(4), 2004.

    International conference presentations 2005-

  • Invited speaker at STOR-i CDT, Lancaster University, June, 2024.
  • Invited speaker at the Winter Conference in Statistics 2024 in Hemavan, March 10-14th.
  • Invited speaker at "Nordic meeting on Insurance Mathematics", Stockholm, May, 2023.
  • Invited speaker at Cramérsällskapets årsmöte , Stockholm, September, 2020 (Zoom webinar).
  • Invited speaker at CASS Business School, London, February, 2020.
  • Invited speaker at Vine Copulas and their Applications , Munchen, July, 2019.
  • Invited speaker at The 23rd International Congress on Insurance: Mathematics and Economics (IME 2019) , Munchen, July, 2019.
  • Speaker at the Colloquium of the International Actuarial Association (IAA) , Oslo, June 2015.
  • Invited speaker at the workshop Recent developments in dependence modelling with applications in finance and insurance, Brussels, May 29th, 2015.
  • Invited speaker at the workshop Workshop On Multivariate Analysis Today (WOMAT), Milton Keynes, May 18-19th, 2015.
  • Invited speaker at the 7th Trondheim Symposium in Statistics , Selbu, Norway, 2014.
  • Invited speaker in the session "Copulas: Past, Present, and Future" at the JSM 2014 in Boston, August 7th, 2014.
  • Invited speaker at the workshop International Workshop on High-Dimensional Dependence and Copulas: Theory, Modeling, and Applications , Beijing, January 4-5, 2014.
  • Invited speaker at the workshop Copulas & extremes , Grenoble, November 19 and 20, 2013.
  • Speaker and session organizer at International Cramér Symposium on Insurance Mathematics , Stockholm, June 11-14, 2013.
  • Invited speaker at the workshop Non-Gaussian Multivariate Statistical Models and their Applications, Calgary, May 19-24, 2013.
  • Invited speaker at Mini symposium on perspective research directions in complex stochastic structures , Eötvös Loránd University, Budapest, March 19, 2012.
  • Invited speaker at the 4th Annual Conference on Extreme Events at the University of Stavanger (UiS) Business School in Stavanger on August 25-26, 2011.
  • Invited speaker at the workshop on Copula Models and Dependence Montréal, June 6-9, 2011.
  • Invited speaker at the 4th Workshop on Vine Copula Distributions and Applications Munich May 11-12, 2011.
  • Invited speaker at the workshop Extreme Value Theory (EVT) & Copula Functions , London, April 5th, 2011 (the workshop was cancelled).
  • Invited speaker at the seminar Climate Policy, Enterprise Risk and Energy Investment, London, November 22-23, 2010.
  • Invited sesssion organizer at The 7th Conference on Multivariate Distributions with Applications , Maresias, Brazil August 8 - 13, 2010.
  • Invited speaker at Workshop on High-dimensional Extremes , Lausanne, 14-18 September 2009,
  • Speaker at 3rd European Risk Conference: "Risk and Accounting", London 3-4 September 2009.
  • Invited speaker at 2nd Vine Copula Workshop, Delft, 16-17 December 2008
  • Invited speaker at Grieg Investor finansseminar, Copenhagen, November 1st, 2008.
  • Invited speaker at the Risk Magazine training course A Quantitative Approach to Calculating and Applying VaR , London, October 3rd, 2008.
  • Invited speaker at Bernoulli Society - IMS WORLD CONGRESS IN PROBABILITY AND STATISTICS 2008 , Singapore, July 14, 2008.
  • Invited plenary speaker at NORDSTAT 2008 in Vilnius, June 18, 2008.
  • Invited to chair the "Statistics in Finance" section at NORDSTAT 2008 in Vilnius, June 17, 2008.
  • Invited speaker at Workshop on Copulae: Theory and Praxis , Berlin Desember 7-8, 2007.
  • Invited speaker at Vine copula workshop , Delft, November 19-20, 2007.
  • Invited speaker at "Alumnidag ved IMF", NTNU, Trondheim, October 26, 2007.
  • Invited speaker at the Energyforum workshop Modelling and pricing in energy markets , Oslo, September 28, 2007.
  • Invited speaker at the Energyforum conference Nordic Modelling & Measuring Energy Risk , Oslo, September 26, 2007.
  • Speaker at the workshop Copulae and multivariate return distributions in finance-Theory, Applications, Opportunities and Problems, Warwick Business School, September 14-15, 2007.
  • Invited speaker at Risk Magazine training course A Quantitative Approach to Calculating and Applying VaR , London, September 4th, 2007.
  • Speaker at Risk Aggregation Seminar, Oslo, August 30, 2007.
  • Speaker at (sfi)2 Workshop on Quantitative Risk Management , Oslo, April 24, 2007.
  • Invited speaker at International Workshop on Computational and Financial Econometrics , Geneve, April 21 2007.
  • Invited speaker at Workshop on Copulas, Lévy processes and Lévy copulas, with applications to financial modelling University of Münich, November 24, 2006.
  • Invited speaker at Econometrics Seminar Swiss Banking Institute , University of Zürich, November 7, 2006.
  • Invited speaker at the conference Risk Magazine's Quant Congress USA 2006 in New York July 12, 2006.
  • Invited speaker at the conference Price Drivers on the Nord Pool Market in Stockholm April 27, 2006.
  • Speaker at International Conference on Finance , København, 2-4 september 2005.
  • Courses, workshops and seminars (most in Norwegian)

  • 3rd Vine Copula Workshop, 15-16 December 2009.
  • Risk Aggregation Seminar, Oslo, August 30, 2007.
  • (sfi)2 Workshop on Quantitative Risk Management , Oslo, April 24, 2007.
  • Statistisk analyse av energipriser
  • Totalrisikomodellering og Basel II
  • Statistiske metoder for analyse av finansielle data
  • Modellering av totalrisiko
  • Fagseminar om copulas
  • Statistisk Analyse av Finansielle Tidsrekker.
  • Supervisions Master

  • Alexander Johansen Ohrt (2023), Probabilistic Tabular Diffusion for Counterfactual Explanation Synthesis, Master of Science, NTNU.
  • Frida Svendal Aase (2023), An Exploration of Shapley Values for Model Interpretability: Providing a Fair and Accurate Explanation of Black Box Models, Master of Science, NTNU.
  • Erik Holst Aasland (2022), Shapley Values for dependent features using Divisive Clustering, Master of Science, NTNU.
  • Arne Rustad (2022),TabGAN: A Framework for Utilizing Tabular GAN for Data Synthesizing and Generation of Counterfactual Explanations, Master of Science, NTNU.
  • Kristine Sivertsen (2016), Interest Rate Models in Solvency II, Master of Science, University of Bergen.
  • Steffen Bjørgum Pedersen (2015), Modellering av sannsynlighet for svindel i bilforsikringskrav, Master of Science, University of Bergen.
  • Linda Mon Xi (2014), Portfolio Optimization with PCC-GARCH-CVar model, Master of Science, University of Bergen.
  • Ingrid Sandvig Thorsen (2013), Modellering av romlig variasjon av frekvenser av vannskader på boliger, Master of Science, University of Bergen.
  • Helene Aardal (2013), DCC-GARCH-modeller med ulike avhengighetsstrukturer, Master of Science, University of Bergen.
  • Lydia Helle (2012), En copula-GLM modell for skadefrekvens og skadestørrelse i bilforsikring, Master of Science, University of Bergen.
  • Elisabeth Orskaug (2009), Multivariate DCC-GARCH models with different error distributions, Master of Science, NTNU.
  • Tormod Sætre (2007), Modeling collateralized debt obligations: A copula approach, Master of Science, NTNU.
  • Henrik Bakken (2005), Copulae: Basic Theory, Goodness-of-Fit Tests and Vines, Master of Science, NTNU.
  • Petter Gravås (2004), Swing Option Valuation Using Monte Carlo Simulations, Master of Science, NTNU.
  • Supervisions PhD

  • Lars Henry Berge Olsen , (2020-2024),Improved Shapley Value Methodology for the Explanations of Machine Learning Models, PhD, University of Oslo.
  • Håvard Kvamme , (2017-2019), Time-to-event Prediction with Neural Networks, PhD, University of Oslo.
  • Rogelio Andrade Macisidor (2016-2020), Machine learning for credit scoring, PhD, University of Tromsø.
  • Ingrid Hobæk Haff (2008-2012), Pair-copula constructions - theory and applications, PhD, University of Oslo.
  • Steffen Grønneberg (2007-2011), New estimation methods and model selection criterias for copulas, PhD, University of Oslo.
  • Daniel Berg (2004-2007), Statistical Analysis of Credit Risk, PhD, University of Oslo.
  • A selection of scientific reports:

  • Aas, Kjersti and Dimakos, Xeni K.: Statistical modelling of financial time series: An introduction. SAMBA/08/04, March, 2004.
  • Aas, Kjersti: Modelling the dependence structure of financial assets: A survey of four copulas. SAMBA/22/04, December, 2004.
  • Aas, Kjersti: Modelling the stochastic behaviour of short-term interest rates: A survey. SAMBA/21/04, September, 2004.
  • Aas, Kjersti: To log or not to log: The distribution of asset returns. SAMBA/03/04, September, 2004.
  • Aas, Kjersti and Hobæk Haff, Ingrid: NIG and Skew Student's t: Two special cases of the Generalised Hyperbolic Distribution. SAMBA/01/05, Januar, 2005.
  • Aas, Kjersti and Hobæk Haff, Ingrid: Modelling a portfolio of financial assets of several different types. SAMBA/24/05, August, 2005.
  • Aas Kjersti: The Basel II IRB approach for credit portfolios: A survey SAMBA/33/05, Oktober, 2005.
  • Aas, Kjersti and Eikvil, Line: Text Categorization: a survey. Technical Report No. 941, Norwegian Computing Center, 1999.
  • All publications:

    Before November 2011

    After November 2011


    Customers:

    I have since 2001 been head of research and marketing for the area : Statistical Analysis for the Finance, Insurance and Commodity markets. at NR. Here is a list of NRs customers in this area.

    Previous research areas:

    I have also long experience in the following fields:

  • Image Analysis: Document analysis, Biometric recognition, Machine vision
  • Video Analysis: Video Surveillance, Tracking, Video indexing
  • Pattern Recognition: Hidden Markov Chain Models
  • Knowledge Mining: Data Mining, Text Mining, Multimedia Mining
  • Intelligent Agents: News filtering agents

  • Why study mathematics? (in Norwegian)

    11 grunner til å velge matte.


    Recreation

    Cross-country skiing
    Mountain climbing