This study aims to develop business failure prediction models using the data of selected firms from ISE markets. The sample data comprise ten selected financial ratios for 27 non-going concerns (failed businesses) and paired 27 going concerns. Two non-parametric classification methods are used in the study: Artificial Neural Networks (ANN) and Decision Trees. The classification results show that there is equilibrium in the classification of the training samples by the models, but ANN model outperform the decision tree model in the classification of the testing samples. Further, the potential usefulness of ANN and Decision Tree type data mining techniques in the analysis of complex and non-linear relationships are observed.
Identifer | oai:union.ndltd.org:METU/oai:etd.lib.metu.edu.tr:http://etd.lib.metu.edu.tr/upload/3/12610621/index.pdf |
Date | 01 June 2009 |
Creators | Tekel, Onur |
Contributors | Sari, Ramazan |
Publisher | METU |
Source Sets | Middle East Technical Univ. |
Language | English |
Detected Language | English |
Type | M.B.A. Thesis |
Format | text/pdf |
Rights | To liberate the content for public access |
Page generated in 0.0015 seconds