Comparing the Bank Failure Prediction Performance of Neural Networks and Support Vector Machines: The Turkish Case

摘要

Experience from the banking crises during the past two decades suggest that advanced prediction models are needed for helping prevent bank failures. This paper compares the ability of artificial neural networks and support vector machines in predicting bank failures. Although artificial neural networks have widely been applied complex problems in business, the literature utilizing support vector machines is relatively narrow and their capability for predicting bank failures is not very familiar. In this paper, these two intelligent techniques are applied to a dataset of Turkish commercial banks. Empirical findings show that although the prediction performance of the two models can be considered as satisfactory, neural networks show slightly better predictive ability than support vector machines. In addition, different types of error from each model also indicate that neural network models are better predictors.Sa?etakIskustvo ste?eno u bankarskoj krizi u posljednja dva desetlje?a upu?uje na potrebu kori?tenja naprednih modela predvi?anja u svrhu prevencije propasti banaka. Ovaj rad uspore?uje sposobnost umjetnih neuronskih mre?a i strojeva s potpornim vektorima da predvide propast banaka. Iako se umjetne neuronske mre?e ?esto koriste za slo?ene probleme u poslovanju, literatura koja spominje strojeve s potpornim vektorima je relativno malobrojna a njihova sposobnost predvi?anja propasti banaka nije previ?e poznata. U ovom radu su ove dvije inteligentne tehnike primijenjene na sklop podataka turskih komercijalnih banaka. Empirijski rezultati pokazuju da iako se predvi?anje dvaju modela mo?e smatrati zadovoljavaju?im, neuronske mre?e pokazuju ne?to bolju sposobnost predvi?anja od strojeva s potpornim vektorima. Osim toga, razli?ite vrste gre?aka u svakom modelu tako?er ukazuju na to da su modeli s neuronskim mre?ama bolji prediktori.