TR
Avrasya Ekonometri �statistik ve Ampirik Ekonomi DergisiYl:2018 Say: 10 Alan: Ampirik ktisat

Volkan SEVN
BANKALARIN BREYSEL KRED TAHSS KARARINA YNELK BR BAYES AI ANALZ
 
Tketicilerin, ihtiyalarn karlayabilmek iin gerekli nakit ihtiyacn salamak amacyla bavurduu zmlerden birisi, bankalardan kredi almaktr. Bu krediler gayrimenkul, ihtiya ve tat kredisi olarak ana grupta toplanabilir. Bankalar, kredi bavurularnn incelenmesi aamasnda, mterilere ait eitli kriterlere gre deerlendirme yapmakta ve kredinin talebinin kabulu ya da reddi ynnde karar vermektedirler. Bu alma, gayrimenkul, ihtiya veya tat kredisine bavuran banka mterilerine ait eitli verilerin, birbirleriye ve bankann karar ile olan ilikilerinin bir Bayes a yoluyla ezamanl analizini sunmaktadr. Yaplan analize gre bankalar ihtiya kredisini daha kolaylkla verme eilimde iken en g alnan kredi tr gayrimenkul kredisi olarak ne kmaktadr. Tat kredisinde ise ya, nemli bir faktr olarak ortaya kmaktadr.

Anahtar Kelimeler: Bayes alar, yapsal renme, gayrimenkul kredisi, ihtiya kredisi, tat kredisi,


A BAYES NETWORK ANALYSIS FOR BANKING'S CONSUMER LOAN ALLOCATION
 
One of the solutions that consumers apply to afford some of their needs is consumer credit. Consumer credits mainly form three groups: Mortgage credits, personal finance credits and vehicle credits. When banks are allocating those credits to their customers they make an evaluation of the customers according to some criteria. At the end of this evaluation they accept or reject the credit demands. This manuscript provides a simultaneous analysis of the variables belonging to the consumer credit customers by means of a Bayesian network. The analysis includes the relationship of the variables with each other and the decision of the bank. According to the analysis it has been detected that banks tend to give the personal finance credit more easily to the all age and occupation groups. The mortgage credit, however, is the most difficult to get. As far as the vehicle credit applications are concerned, age appears to be an important factor to get the credit.

Keywords: Bayesian networks, structural learning, mortgage credit, personal finance credit, vehicle credit


Detay

ÇERK