Neural network prediction analysis: The bankruptcy case

Artificial neural networks in bankruptcy prediction: General framework and cross-validation analysis

“…For example, Raghupathi et al [48] In order to detect maximal dierence between bankrupt and nonbankrupt ®rms, many studies employ matched samples based on some common characteristics in their data collection process. Characteristics used for this purpose include asset or capital size and sales [19,36,63], industry category or economic sector [48], geographic location [55], number of branches, age, and charter status [61]. This sample selection procedure implies that sample mixture ratio of bankrupt to nonbankrupt ®rms is 50% to 50%.…”

“…Results show that the performance of dierent classi®ers depends on the proportions of bankrupt ®rms in the training and testing data sets, the variables used in the models, and the relative cost of Type I and Type II errors. Boritz and Kennedy [8,9] Leshno and Spector [36] evaluate the prediction capability of various ANN models with dierent data span, neural network architecture and the number of iterations. Their main conclusions are (1) the prediction capability of the model depends on the sample size used for training; (2) dierent learning techniques have signi®cant eects on both model ®tting and test performance; and (3) over®tting problems are associated with large number of iterations.…”

Vol. 2, No. 4, October 2009. All in one file.

“…Artificial neural network is extensively being applied in predicting bankruptcy. Leshno & Spector (1996) have compared artificial neural network with multivariate discriminant analysis and logistic regression in their study on bankruptcy using a limited number of firms. Prediction capabilities of artificial neural network are found to be more accurate than the classical discriminant analysis and logistic regression.…”

“…For example, artificial neural network is compared with multiple linear regression (Nguyen &Cripps, 2001 andArulsudar, Subramaniam &Murthy, 2005), discriminant analysis and logistic regression (Leshno & Spector, 1996), decision trees and logistic regression (Delen, Walker & Kadam, 2004), stepwise regression and ridge regression (Chokmani,Quarda, Hamilton, Hosni & Hugo, 2008), logistic regression (Zhang, Hu, Patuwo & Indro, 1997). The artificial neural network has outperformed the traditional methods in all of these studies.…”

Predicting bankruptcy using neural networks and other classification methods: The influence of variable selection techniques on model accuracy

“…Lee et al [60] Variables used in study by Altman [2] MLP-LM -SOM Leshno and Spector [61] Judgment and correlation test applied to variables used in previous studies and variables used in study by Altman [2] MLP-BP -MLP-?…”