http://hdl.handle.net/20.500.12627/39 2026-05-01T03:38:47Z http://hdl.handle.net/20.500.12627/167874 Robust Mahalanobis Distance based TOPSIS to Evaluate the Economic Development of Provinces Yıldırım, Bahadır Fatih; Yorulmaz, Özlem; Kuzu Yıldırım, Sultan In this paper, 81 Turkish provinces with different development levels were ranked using the TOPSIS method. To evaluate the ranking with TOPSIS, we presented an improvement to Mahalanobis distances, by considering a robust MM estimator of the covariance matrix to deal with the presence of outliers in the dataset. Additionally, the homogenous subsets, which were obtained from the robust Mahalanobis distance-based TOPSIS were compared with robust cluster analysis. According to our findings, robust TOPSIS-M scores reflect the inter-class differences in economic developments of provinces spanning from the extremely low to the extremely high level of economic developments. Considering indicators of economic development, which are often used in the literature, İstanbul ranked first, Ankara second, and İzmir third according to the Robust TOPSIS-M method. Moreover, with the Robust Cluster analysis, these provinces were diagnosed as outliers and it was seen that obtained clusters were compatible with the ranking of Robust TOPSIS-M. 2021-07-01T00:00:00Z http://hdl.handle.net/20.500.12627/615 Time Period Based COPRAS-G Method: Application on the Logistics Performance Index Adıgüzel Mercangöz, Burcu; Yıldırım, Bahadır Fatih; Kuzu Yıldırım, Sultan Background: Logistics is vital for the trades of countries. The inputs such as raw materials and energy that is needed for production and also the outputs of these processes are transported and distributed effectively as a result of an efficient logistics process. In order to measure the logistics performance of countries, The World Bank (WB) is publishing an index entitled Logistics Performance for every two years. Methods: The main value of this study is to provide logistics performance scores of the selected countries for a selected time period. Thus, periodic evaluations can be done for a selected time period. The grey numbers are used for determining a new dataset for a time period and implement to Complex Proportional Assessment of Alternatives (COPRAS) method. 28 European Union (EU) member states plus 5 EU Candidate Countries are ranked by using the COPRAS-Grey (COPRAS-G) method according to their logistics performance scores. In order to see if the ranking calculated by COPRAS-G is representing the past index data, the bilateral comparisons of the rankings are investigated by using the Spearman Rank and Kendall’s Tau Correlation methods. Results: The results showed that the dataset obtained by using grey numbers represent the LPI scores of the countries for the selected time period. Although there are slight differences between the Spearman and Kendall correlation coefficients, the ultimate result is the same. The ranking calculated by COPRAS-G has the strongest relationship with all rankings published by WB. Conclusions: By using the grey numbers combined with the COPRAS-G method, the LPI of Countries can be evaluated for a time period. 2020-01-01T00:00:00Z http://hdl.handle.net/20.500.12627/614 Kredi Kartı Platformlarının Sezgisel Bulanık TOPSIS Yöntemi Kullanılarak Değerlendirilmesi Yıldırım, Bahadır Fatih Türkiye’de kredi kartı pazarında artan penetrasyon ve rekabetle birlikte verim-lilik önem kazanmıştır. Yatırımların geri dönüş süresinin çok uzun vadeli olması bankaları işbirliğine yöneltmiştir. Bankalar arası işbirliğinin bir sonucu olarak kredi kartı markaları birden fazla bankanın paylaştığı ve müşterek kullandıkları platform-lara dönüşmüştür. Bu çalışmada Türkiye’de faaliyet gösteren 7 çatı kredi kartı plat-formu Sezgisel Bulanık TOPSIS yöntemi kullanılarak değerlendirilmiş, belirlenen krit-erlere göre uzman görüşüne dayalı bir skorlama ile sıralanmıştır. Analiz sonuçları kredi kartı markalarının mevcut pazar payları ile kıyaslanmış, elde edilen bulguların tutarlı olduğu saptanmıştır. 2019-01-01T00:00:00Z http://hdl.handle.net/20.500.12627/613 Network Risk Modelling via Warped Gaussian Processes Adiguzel Mercangoz, Burcu; W. Peters, Gareth Many problems in Operational Research domains involve the study of risk management on a constrained topology of some form of physical network such as logistic networks or intangible networks such as banking financial networks and distributed ledger networks in modern DeFi movements. In this paper we explore an emerging class of machine learning models that can flexibly model and assess risk on a graphical or network based topology. We will focus primarily on the use of undirected graphical structures and we will develop a stochastic model representa- tion that is able to model dynamic multivariate processes on a graphical topology conditional on exogenous observable covariates. This will be achieved by extending classical Gaussian processes regressions models in machine learning to non- stationary, warped Gaussian process regression models. Since we are focused on risk management aspects of processes restricted to graphs, we note that extreme events often display heterogeneity (i.e., non-stationarity), varying continuously with a number of covariates. In the framework we develop we will be able to study and explain such extreme joint variations through the regressions structures developed and the covariance operators characterizing the process. Having proposed a class of such warped Gaussian process network regression models, we will then study Bregman super-quantiles on such networks that will allow us to develop a class of network based coherent risk measures which have the added advantage of being sub-additive, allowing one to aggregate of local graphical cliques to understand local risk behaviours. 2019-09-19T00:00:00Z