Thesis supervisor: Edith Alice Kovács
Location of studies (in Hungarian): Department of Differential Equations, Institute of Mathematics, BME Abbreviation of location of studies: BME
Description of the research topic:
Copulas are a useful tool in modeling multivariate probability distribution which play an important role in generating scenarios for different fields: Finance: estimate the credit risk and the market risk, Insurance, Hydrology. As Fisher (1997) notes in the Encyclopedia of Statistical Sciences : “Copulas are of interest to statisticians for two main reasons: First, as a way of studying scale-free measures of dependence; and secondly, as a starting point for constructing families of bivariate distributions, [...]” For the fitting of two dimensional copulas to sample data there are a lot a very good algorithms and programs.
In higher dimensions often appear different types of dependences between the random variables involved. For this aim there were introduced the regular vine copulas, which can model many types of dependences between different pairs of random variables. However this is also a drawback since the number of parameters becomes very large, as the dimension of the multivariate random variables grows. To reduce this large number of parameters, the Truncated vine copulas and the Cherry-tree copulas were introduced. The candidate is supposed to do research in developing the modeling of cherry tree copulas and truncated vines.
The candidate is supposed to have some routine in Probability Theory. Furthermore, one of the tasks of the candidate will be deepening this knowledge by adding Copula Theory and some parts of Information Theory.
Required language skills: English Further requirements: To graduate introductory courses from Algebra, Probability Theory
Number of students who can be accepted: 1
Deadline for application: 2018-05-31
2024. IV. 17. ODT ülés Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).
Login
