Department of Information Technology

Computational Finance

We are interested in the numerical pricing of financial derivatives. The numerical approximation techniques used are adaptive finite difference methods and radial basis functions. We consider pricing of both European and American options as well as some exotic options such as Turbo warrants, bubbles and the term structure equation. The whole range of the solution process is considered, from the numerical discretization to the solution of the resulting linear systems using iterative methods and preconditioning.

We have also in 2013 started research on model calibration and parameter estimation for financial models. This is needed for model consistent risk management of financial derivatives (hedging, risk assessment like "conditional value at risk" computations), something which is emphasized by new stricter regulations. Our idea is to combine advanced numerical methods with statistics in order to propose new statistical tools that are very much needed, especially in light of the rapid increase of the volume of the financial derivative markets, leading to huge volumes of trading data and large numbers of transactions in need of accurate statistical information.

Since 2014 we are leading the large international collaboration project BENCHOP. The purpose and aim of this project is to provide sets of benchmark problems that can be used for comparison and evaluation of methods and to serve as a take off for future development of methods in option pricing. The project has so far resulted in two publications

Activities organized by the group

  • A workshop on Mathematical and Numerical Modeling in Finance was organized by Josef Höök, Elisabeth Larsson and Lina von Sydow June 9-11 2014 at Institut Mittag-Leffler in Djursholm. It was well attended by people from academia and financial institutions. There were research presentations with a mix of applied mathematics and numerical methods in the morning. New problems were discussed in the afternoon sessions, which resulted in at least one new major collaboration project.
  • May 8-9 2014, Slobodan Milovanovic and Victor Shcherbakov accompanied by Elisabeth Larsson took part in the organization of and attended Postgraduate Workshop on Approximation Theory (with emphasis on Financial Applications) at Birkbeck, University of London. Doctoral students and researchers from Birkbeck, Leicester University, and Uppsala University had a chance to share their research results and exchange ideas amongst a larger audience. The broader aim of the workshop was to explore potential collaborative projects in this field for the future.
  • On October 8 2013 the group organized a Workshop in Computational and Mathematical Finance. The workshop was very successful with many interesting presentations in the field that attracted 50 participants from business and academia. We plan to make this a recurrent event.

Refereed publications

PhD Theses

Reports (nonoverlapping with refereed publications)

Supervised Master thesis projects

Supervised Bachelor thesis projects

Supervised projects in project course in Scientific Computing

Updated  2019-03-08 14:49:18 by Lina von Sydow.