Vinnere av aktuarprisen 2022

Prisen deles ut av Den Norske Aktuarforening, og er på totalt 25.000 kr. I år ble det to prisvinnere: Sandra Vervik Heimsæter og Åsmund Hausken Sande.

Sandra Vervik Heimsæter, ble uteksaminert fra Universitetet i Bergen i februar 2021 og tildelt aktuarkompetansebevis mars 2021. Oppgaven omhandler en ny metode innen statistisk læring, som er nyttig for å gjøre prediksjoner når store datamengder er tilgjengelig, som er tilfelle i et forsikringsselskap. Veiledere: Førsteamanuensis Yushu Li, Stipendiat Ingvild M. Helgøy. Sammendrag av oppgaven: The thesis  developed an algorithm which can simultaneously achieve sample and feature selection when facing big data in supervised learning. This parametric Bayesian regression learning algorithm is based on a well known Bayesian sparse learning method: the Relevance Vector Machine (RVM). The deduction of the algorithm is inspired by, the probabilistic feature selection and classification vector machine (PFCVM), which is a simultaneous sample and feature selective extension of the RVM classification model. The developed method in the thesis is called the dimensionality reducing relevance vector machine (DRVM), and it performs simultaneous feature and sample selection in the regression case. Empirical examples are also included using standard benchmark data set. Oppgaven er tilgjengelig på: https://bora.uib.no/bora-xmlui/handle/11250/2737181.

Sandra jobber nå som prisaktuar i Tryg Forsikring. Hun ble medlem av Den Norske Aktuarforening i november 2021.

Åsmund Hausken Sande: Oppgaven presenterer en utvidet setting for stokastiske renter for livsforsikringsavtaler. Veileder: David R. Banos. Tittel: Prospective Reserves of Life Insurance Policies under the Heath-Jarrow-Morton Framework. Sammendrag av oppgaven: In this thesis we consider a general stochastic interest rate under the HJM (Heath-Jarrow-Morton) framework. We further present a general model for the pricing of life insurance policies allowing for a wider range of stochastic policy functions than previously done within the HJM framework. This is carried out by modelling them under a general financial market model with Gaussian noise. Furthermore, we develop standard pricing formulas based on financial arbitrage methods for both current time and future time-points. It is worth noting that these equations are contingent on formulas pricing the instantaneous values of the policy functions as financial claims. Lastly, we give an example where the theory is applied to exactly evaluate the price of reserves within a new theoretical pension scheme with stochastic policy functions tied to the interest rate. As a part of this example we develop small generalizations of some financial pricing formulas for call and digital options on zero-coupon bonds. In order to rigorously justify these results the thesis covers a large amount of background material. This includes measure and probability theory, as well as using these to introduce important concepts in interest rate, finance and classical insurance theory. Oppgaven er tilgjengelig på: https://www.duo.uio.no/handle/10852/86894 

Åsmund er nå stipendiat – Stokastisk, finans og risiko, Universitetet i Oslo.

Takk til priskomiteen bestående av Patrick Kakunce, leder av Utdanningskomiteen, professor Bård Støve, Universitetet i Bergen og førsteamanuensis David R. Banos, , Universitetet i Oslo

Den Norske Aktuarforening gratulerer!