Αξιολόγηση πιστοληπτικής ικανότητας δανειοληπτών μέσω στατιστικών τεχνικών
Creditworthiness assessment of borrowers through statistical techniques
View/
Keywords
Πιστοληπτική ικανότητα ; Αποπληρωμή δανείων ; Πιθανότητα αθέτησης ; Στατιστική ανάλυση ; Credit scoring ; Πιστωτικός κίνδυνοςAbstract
This thesis focuses on the analysis and prediction of loan repayment delays using statistical methods, without the use of machine learning techniques. The main objective is to investigate the factors affecting credit risk and to develop a simple borrower risk assessment model. The analysis is based on the German Credit Dataset, which includes data for 1,000 borrowers along with various attributes related to their financial and demographic characteristics, as well as loan features. The methodology relies on descriptive statistics, correlation analysis, and probability of default (PD) estimation. The results indicate that loan duration and loan amount are key factors influencing the probability of default, while age has a less significant impact. Furthermore, a risk segmentation model was developed, classifying borrowers into low, medium, and high-risk categories, revealing substantial differences in default rates across these groups. The findings demonstrate that statistical analysis can serve as an effective and transparent tool for credit risk evaluation, providing valuable insights for decision-making in the banking sector.
Postgraduate Studies Programme
Πληροφοριακά Συστήματα και ΥπηρεσίεςDepartment
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών ΣυστημάτωνNumber of pages
66Language
GreekCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα