izpis_h1_title_alt

Modeliranje izgube ob dogodku neplačila stanovanjskih kreditov in potencialne izboljšave : magistrsko delo
ID Skočir, Matic (Author), ID Bernik, Janez (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (2,86 MB)
MD5: 6D0B658590AC9A8D235E6F2FE7653998

Abstract
Upravljanje kreditnega tveganja je ena poglavitnih nalog bank, saj zagotavlja stabilnost poslovanja, hkrati pa omogoča pridobivanje ključne prednosti v bančnem sistemu. V magistrskem delu je opisan proces modeliranja izgub ob dogodku neplačila stanovanjskih kreditov. V delu je predstavljena konkretna rešitev, konstruirana na eni izmed slovenskih bank, s pomočjo zgodovinskih podatkov o izterjavi poslov. Nadalje so predstavljene potencialne izboljšave ocenjevanja izgube. Prikazane so izboljšave pri vpeljavi makroekonomskih kazalnikov v model in izboljšave pri vpeljavi metod strojnega učenja. S pomočjo mer kakovosti so vse tehnike strojnega učenja v delu ovrednotene. V zadnjem delu je predstavljen proces oblikovanja rezervacij stanovanjskih kreditov, ki sledi izračunu izgube ob dogodku neplačila.

Language:Slovenian
Keywords:kreditno tveganje, izguba ob dogodku neplačila, krivulja preostale pričakovane izterjave, strojno učenje
Work type:Master's thesis/paper
Typology:2.09 - Master's Thesis
Organization:FMF - Faculty of Mathematics and Physics
Year:2018
PID:20.500.12556/RUL-103876 This link opens in a new window
UDC:519.8
COBISS.SI-ID:18453593 This link opens in a new window
Publication date in RUL:28.09.2018
Views:1021
Downloads:205
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Modelling loss given default of housing loans with potential improvements
Abstract:
Managing credit risk is one of main tasks banks encounter. It ensures business stability and presents opportunity of gaining key advantage in banking system. The focus of the master thesis is on describing the process of modelling of loss given default of housing loans. Actual solution which was built using historical data of debt collection is presented. Afterwards, the potential improvements of assessing severity of loss are introduced. Improvements consist of introducing macroeconomic factors into the model and using of machine learning methods. All the machine learning techniques, which were used for modelling, are assessed by performing quality measures. Lastly, construction of provisioning process is introduced.

Keywords:credit risk, loss given default, remaining recovery curve, machine learning

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back