Full-text resources of PSJD and other databases are now available in the new Library of Science.
Visit https://bibliotekanauki.pl
Preferences help
Polish version English version Language
enabled [disable] Abstract
Number of results

Results found: 4

Number of results on page
first rewind previous Page / 1 next fast forward last

Search results

help Sort By:

help Limit search:
first rewind previous Page / 1 next fast forward last
1
Content available remote

Propensity Score Matching and Its Application to Risk Drivers Detection in Financial Setting

100%
Karwański M., Grzybowska U.
Acta Physica Polonica A
|
2016
|
vol. 129
|
issue 5
945-949
EN
In credit risk scoring models are used as a tool to evaluate the level of risk associated with applicants or customers. The aim of these models is not only to estimate the probability that the client will not be able to fulfill his financial commitments but also to identify and estimate the risk drivers i.e., client attributes that are responsible for risk occurrence. Unfortunately, scoring models are built based on historic data stored by bank over the clients. Selection of clients is not random. This leads to systematic errors. Therefore one seeks methods that allow for a model correction that enables application of statistical inference. Quasi-experimental designs are practical solutions to this dilemma. One of such methods is propensity score matching. Propensity score matching allows also for detecting risk drivers that are independent of borrowers attributes, e.g., triggered by various bank strategies. The aim of our research is to apply propensity score matching methodology to identify these risk drivers in credit risk that could not be detected e.g., by regression models.
2
Content available remote

Modeling Correlations in Operational Risk

100%
Karwański M., Grzybowska U.
Acta Physica Polonica A
|
2018
|
vol. 133
|
issue 6
1402-1407
EN
The key demand for banks' economic capital methodology is to ensure that the model covers all relevant sources of risk in the right way. Operational risk models treat the arising losses as stochastic variables. One of the problems encountered in modeling is the need of taking into account correlations between events. It is possible to build models for correlated events based on copula functions. But the problem is that the losses are related to isolated events and simple applications of copulas are not allowed. The authors present a new algorithm that shows a modified application of copulas to calculating operational risk. The calculations were done on real data that allows for examining the correlation impact on risk measurement. As an additional evaluation of the algorithm a reference model based on the Pareto-Lévy copulas was used.
3
Content available remote

Application of Data Envelopment Analysis to Calculating Probability of Default for High Rated Portfolio

100%
Grzybowska U., Karwański M.
|
EN
The aim of our research is to propose a method of rating companies which is based on efficiency measure given by Data Envelopment Analysis (DEA). Proper rating of borrowers is an essential requirement of PD estimation. The difficulty in DEA application is the selection of input and output from the set of indicators describing evaluated objects, which is usually based on expert knowledge. Therefore we apply random forests and gradient boosting to select financial indicators used by the DEA approach and to obtain a ranking of companies needed for PD estimation.
4
Content available remote

Examples of Migration Matrices Models and their Performance in Credit Risk Analysis

81%
Grzybowska U., Karwański M., Orłowski A.
Acta Physica Polonica A
|
2012
|
vol. 121
|
issue 2B
B-40-B-46
EN
Credit risk models used in banks are based on probability models for occurrence of default. A vast class of the models used in practice (e.g., Credit Metrics) is based on the notion of intensity. In 1997 Jarrow applied Markov chain approach to analyze intensities. The key problem that arises is the selection of appropriate estimators. Within the Markov approach among the most frequently used estimators of a migration matrix are cohort and duration estimators. Migration matrices can also be obtained with help of statistical longitudinal models (GLMM) in which states (rating classes) in discrete time points are regarded as matched pairs. In this paper we compare Markov chain models and GLMM models and the influence of their application on bank portfolio evaluation.
first rewind previous Page / 1 next fast forward last
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.