Institutional Repository of Vilnius University
Įsipareigojimų neįvykdymo dažnio prognozavimas Baltijos šalių paskolų portfeliams /
Abstract
dc:descriptionIssuing loans is one of the most profitable, yet one of the riskiest activities for banks. The key factor ensuring banks' profitability and operational efficiency is accurate and appropriate monitoring and management of credit risk, as well as the provision of sufficient capital for unexpected losses. One of the main criteria for credit risk is the borrower's probability of default. Accordingly, a risk class rating is assigned and the bank unit is obliged to set aside a certain amount of capital buffer for the overall portfolio under the policy of International Financial Reporting Standards (IFRS). The main aim of this research paper is to investigate whether changes in macroeconomic factors have a statistically significant impact on default rates, taking into account the borrower's initial risk class. In other words, changes in macroeconomic indicators can lead to the occurrence of client defaults. The initial hypothesis is that clients with the lowest risk class are the most exposed to changes and therefore it is of utmost importance to properly assess the relevant external factors before assigning the credit risk class. Multiple linear regression models, dependent variable transformation, and model validation are used to achieve the goal and to confirm or deny the hypothesis. The second part of the study is to develop and compare default rate forecasts between two models, using time series models. The first model is constructed using only past, historical values of default events, while the second model includes the growth rate of macroeconomic factors as exogenous regressors.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Alčauskytė, Mintė,
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:ELABAETD192845888&prefLang=en_US