Predicting business failures in non-financial turkish companies
Author
Okay, Kaan
Advisor
Esmer, Burcu
Date
2015Publisher
Bilkent University
Language
English
Type
ThesisItem Usage Stats
136
views
views
52
downloads
downloads
Abstract
The prediction of corporate bankruptcies has been widely studied in the finance
literature. This paper investigates business failures in non-financial Turkish companies
between the years 2000 and 2015. I compare the accuracies of different
prediction models such as multivariate linear discriminant, quadratic discriminant,
logit, probit, decision tree, neural networks and support vector machine
models. This study shows that accounting variables are powerful predictors of
business failures one to two years prior to the bankruptcy. The results show
that three financial ratios: working capital to total assets, net income to total
assets, net income to total liabilities are significant in predicting business failures
in non-financial Turkish companies. When the whole sample is used, all five
models predict the business failures with at least 75% total accuracy, where the
decision tree model has the best accuracy. When the hold-out samples are used,
neural networks model has the best prediction power among all models used in
this study.
Keywords
multivariate linear discriminant modelquadratic discriminant model
logit model
probit model
decision tree model
neural networks model
support vector machines
business failures
bankruptcy prediction
financial ratios
Permalink
http://hdl.handle.net/11693/30046Related items
Showing items related by title, author, creator and subject.
-
Interaction-based feature-driven model-transformations for generating E-forms
Tekinerdoǧan, Bedir; Aktekin, N. (ACM, 2009-10)One of the basic pillars in Model-Driven Software Development (MDSD) is defined by model transformations and likewise several useful approaches have been proposed in this context. In parallel, domain modeling plays an ... -
Architecture framework for mapping parallel algorithms to parallel computing platforms
Tekinerdogan, Bedir; Arkin, E. (CEUR-WS, 2013)Mapping parallel algorithms to parallel computing platforms requires several activities such as the analysis of the parallel algorithm, the definition of the logical configuration of the platform, and the mapping of the ... -
Influence of phase function on modeled optical response of nanoparticle-labeled epithelial tissues
Cihan, C.; Arifler, D. (2011)Metal nanoparticles can be functionalized with biomolecules to selectively localize in precancerous tissues and can act as optical contrast enhancers for reflectance-based diagnosis of epithelial precancer. We carry out ...