20.500.12556/DKUM-48207 VPLIV MAKROEKONOMSKEGA OKOLJA NA VERJETNOST NEPLAČILA ZA PORTFELJ SEKTORJA GOSPODINJSTEV Impact of Macroeconomic Environment on Probability of Default for Porfolio of Household Sector V magistrskem delu obravnavamo področje kreditnega tveganja sektorja gospodinjstev, ki se v zadnjih letih sooča s turbulentnimi tržnimi razmerami. Raziskavo smo izvedli na podlagi pridobljenih podatkov slovenske poslovne banke na področju kreditojemalcev dolgoročnih kreditov. Baza podatkov obsega naključne kreditojemalce, ki so imeli najet dolgoročni kredit v letu 2010 in 2013. Vsakega od teh kreditojemalcev smo spremljali tekom posameznega leta in pri tem zbirali podatke o številu dni zamude, kar služi za razvrstitev ali je kreditojemalec slab ali dober. Na podlagi statistično značilnih pojasnjevalnih spremenljivk smo za obe opazovani obdobji sestavili logit model z najvišjo napovedno močjo. Model nam je dal podatke o tem katere spremenljivke imajo pozitiven oz. negativen vpliv na verjetnost, da bo kreditojemalec slab oz. dober ter podatke o napovedni moči modela. Na koncu smo sestavili logit model za testiranje strukturnih prelomov. Pri gradnji modela smo vključili slamnate pojasnjevalne spremenljivke. Na podlagi statistično značilnega regresijskega koeficienta pojasnjevalne spremenljivke »dummy« smo potrdil zastavljeno hipotezo in sicer da spremembe makroekonomskega okolja povzročajo strukturne prelome v modelu neplačil sektorja gospodinjstev. The master´s thesis presented credit risk of the household sector, which has facing a turbulent market conditions in recent years. The survey was carried out on the basis of information obtained from Slovenian commercial bank on the field of borrowers of long-term loans. The database comprises a random borrowers who have hired long-term loan in 2010 and 2013. Each of these borrowers were monitored over the course of each year, while collecting data on the number of days past due which is used to classify whether the borrower is bad or good. Based on the statistically significant explanatory variables we assembled logit model with the highest predictive power for both observed periods. Model gave us information about which variables have a positive or. negative impact on the probability that a borrower are good or bad and data of the predictive power of the model. At the end of master´s thesis we presented a logit model for testing structural breaks. We have included dummy explanatory variables when building the model. Based on statistically significant regression coefficient of the dummy explanatory variable we confirmed our hypothesis that changes in the macroeconomic environment created structural breaks in the default model of the household sector. kreditno tveganje prebivalstva logit model verjetnost neplačila strukturni prelomi. credit risk of households logit model probability of default structural breaks. true true false [B. Kenda] Slovenski jezik Angleški jezik Magistrsko delo 2015-05-30 10:28:13 2016-01-22 11:52:38 2022-05-24 03:06:35 2021-05-08 03:35:02 2015 0 Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 336.77 12199964 URN:SI:UM:DK:QKTXWNS7 MAG_Kenda_Boris_2015.pdf MAG_Kenda_Boris_2015.pdf 1 CEC0CEE686E6D8D0B91925AF80D16E52 f1acfa93abc0ec57ccd00112632d9162cfc7cc57b3b9ba2337fd44aeea75892b c6ed9944-7c0b-11eb-bb7a-00155d0001ca https://dk.um.si/Dokument.php?lang=slv&id=82202 Ekonomsko-poslovna fakulteta 0 0 0