Institutional Repository of Vilnius University
Nuostolių dėl įsipareigojimų nevykdymo matematinis modeliavimas /
Abstract
dc:descriptionMathematical modeling of loss given default As the people are willing to take more loans for their individual use, the banks have to estimate the loan conditions more carefully. The level of consumptions is growing rapidly and not all people are evaluating their financial ability to pay the loan. That is why a lot of loans are not being returned and the banks are facing great loss over the time. To tackle this particular problem, banks have to evaluate their client‘s creditworthiness now, more careful then ever. And the main object regarding this evaluation is loss given default. The importance of knowing what loss given default can be for a particular loan is almost crucial, therefore, this paper introduced the main knowledge of what loss given default is. The first part of the paper presents the main information about all the necessary concepts used in the theory of loan evaluation. This part includes the information and definitions of these main characteristics: the probability of default, exposure at default, loss given default, expected loss, unexpected loss and value at risk. Of course, there are some other definitions that are used as an addition to the main objects, but are as important as the definitions mentioned before. The second part of this paper includes propositions for describing the formulas, used in the loan portfolio evaluation. It presents the main formulas of expected loss and unexpected loss in the loan portfolio evaluation and for the clarity of the formulas, some of the propositions are given with proof. Moreover, this chapter has the explanation of the Monte Carlo simulation and the presentation of the analytical approximation for the loss distribution. The last part displays the practical calculations of the paper. In this sector the specific loan portfolio was taken. Because of the size of portfolio, the subset was taken to evaluate the portfolio. In order to check fulfillment of the taken subset, the portfolio was grouped according to loan debts and grouped again according to loan loss. Then the distribution of portfolio and of the subset was compared. The distribution in the generated groups were the same with the error percentage of less than 0,5%. The subset had the same distribution in generated debts‘ and loss‘ groups. That is why, the loss distribution of the subset had the same shape and other indicators, such as the intervals of growing and decreasing. Using the Monte Carlo simulation, the histogram of the portfolio‘s loss given default was done. After this step, using the knowledge of analytical approximation, it was tested whether loss given default distribution match the beta distribution. The average and variance of the simulated loss given default was calculated and because of these characteristics the parameters of beta distribution where founded.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Gumbakytė, Gabija,
- Contributors dc:contributor
-
- Grigutis, Andrius
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:ELABAETD35374887&prefLang=en_US