Marek Karwanski

Dr. Marek Karwanski

orcid

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 January 2022 in Entropy
Verified
Journal Article
Open access
PDF file
UGUrszula Joanna Grzybowska
Urszula Joanna Grzybowska
MKMarek Karwanski
Marek Karwanski
One of the goals of macroeconomic analysis is to rank and segment enterprises described by many financial indicators. The segmentation can be used for investment strategies or risk evaluation. The aim of this research was to distinguish groups of similar objects and visualize the results in a low dimensional space. In order to obtain clusters of similar objects, the authors applied a DEA BCC model and archetypal analysis for a set of companies described by financial indicators and listed on the Warsaw Stock Exchange. The authors showed that both methods give consistent results. To get a better insight into the data structure as well as a visualization of the similarities between objects, the authors used a new approach called the PHATE algorithm. It allowed the results of DEA and archetypal analysis to be visualized in a low dimensional space.

1 December 2021 in Resources Policy
Verified
Journal Article
Closed access
BBBolesław Borkowski
Bolesław Borkowski
MKMonika Krawiec
Monika Krawiec
MKMarek Karwanski
Marek Karwanski
WSWiesław Szczesny
Wiesław Szczesny
YSYochanan Shachmurove
Yochanan Shachmurove

12 March 2024 in Medical Science Monitor
Verified
Journal Article
Open access
WKWojciech Klimm
Wojciech Klimm
KSKatarzyna Szamotulska
Katarzyna Szamotulska
MKMarek Karwanski
Marek Karwanski
ZBZbigniew Bartoszewicz
Zbigniew Bartoszewicz
WWWojciech Witkowski
Wojciech Witkowski
Stanisław NiemczykStanisław Niemczyk
Stanisław Niemczyk
Background: Acute kidney injury (AKI) is a common and serious complication after massive burn injury. One of the postulated etiologies is destruction of the extracellular matrix of nephrons, caused by a local imbalance between matrix metalloproteinases (MMPs) and specific inhibitors. The aim of this study was to analyze the dynamics of tissue inhibitors of metalloproteinases (TIMPs) during the first 5 days after massive thermal injury and the relationship with the risk of AKI. Material/Methods: Thirty-three adults (22 men, 11 women) with severe burns were enrolled in the study. The values of TIMPs 1 to 4 were measured in blood serum and urine using the multiplex Luminex system. The associations between TIMPs and the risk of AKI were analyzed by using the generalized linear mixed models for repeated measurements. Results: Significant changes in serum and urine activities of TIMPs were confirmed, especially during the first 2 days after burn injury. Almost half of patients presented renal problems during the study. Significant differences between values of TIMPs in AKI and non-AKI status were also observed. However, a significant relationship between concentration of TIMPs and risk of AKI was confirmed only for urine TIMP-1 and serum TIMP-3. Conclusions: The evaluation of TIMPs in the early stage after burn injury has potential benefits. The important roles of urine TIMP-1 and serum TIMP-3, as novel markers of the risk of AKI development, were confirmed. Other parameters require further analysis.

18 June 2024 in Metody Ilościowe w Badaniach Ekonomicznych
Verified
Journal Article
Open access
PDF file
UGUrszula Joanna Grzybowska
Urszula Joanna Grzybowska
MKMarek Karwanski
Marek Karwanski
In line with new international financial supervision directives (IFRS9), banks should look at a new set of analytical tools, such as machine learning. The introduction of these methods into banking practice requires reformulation of business goals, both in terms of the accuracy of predictions and the definition of risk factors. The article compares methods for selecting variables and assigning "importance" in statistical and algorithmic models. The calculations were carried out using the example of financial data classification for loan default. The effectiveness of various machine learning algorithms on selected sets of variables was compared. The results of the analyzes indicate the need to revise the concept of the "importance" of a variable so that it does not depend on the structure of the model.


2 January 2024 in Metody Ilościowe w Badaniach Ekonomicznych
Verified
Journal Article
Open access
PDF file
UGUrszula Joanna Grzybowska
Urszula Joanna Grzybowska
MKMarek Karwanski
Marek Karwanski
Zgodnie z nowymi dyrektywami międzynarodowego nadzoru finansowego (MSSF9) banki powinny przyjrzeć się nowemu zestawowi narzędzi analitycznych, takich jak uczenie maszynowe. Wprowadzenie tych metod do praktyki bankowej wymaga przeformułowania celów biznesowych, zarówno w zakresie trafności przewidywań, jak i definicji czynników ryzyka. W artykule porównano metody selekcji zmiennych i przypisania „ważności” w modelach statystycznych i algorytmicznych. Obliczenia przeprowadzono na przykładzie klasyfikacji danych finansowych. Na wybranych zbiorach zmiennych porównano skuteczność różnych algorytmów uczenia maszynowego. Wyniki analiz wskazują na potrzebę rewizji koncepcji „ważności” zmiennej, tak aby nie była ona zależna od struktury modelu.