Institutional Repository of Vilnius University
Įsipareigojimų neįvykdymo tikimybės vertinimas modelyje su sisteminiu faktoriumi taikant Bajeso metodą /
Abstract
dc:descriptionThis thesis examines the estimation of the probability of default using the Bayesian method in a model with a systemic factor. The probability of default is treated as an unknown parameter to which a prior distribution is assigned. Taking into account the event that the number of defaults in the portfolio does not exceed a predetermined threshold k, the posterior distribution of this parameter is derived. The thesis first discusses the basic binomial model and then extends it by introducing one systemic factor, which describes the influence of common economic conditions on borrowers. In the practical part, three prior distributions are compared: the Beta distribution, the logit-normal distribution, and a pessimistic prior, which assigns greater weight to higher values of the probability of default. The model is applied to three risk classes. The main focus is placed on the supremum of the probability of default, determined using an analogue of the Type I error. The obtained results show that the choice of the prior distribution has a greater impact in higher-risk classes, while the pessimistic prior produces the most conservative upper bounds in all analysed cases.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Mazur, Gabriela Jolanta,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308327209&prefLang=en_US