Application of Neural Networks to Bond Rating and House Pricing

Abstract

Feed forward neural networks receive a growing attention as a data modelling tool in economic classification problems. It is well-known that controlling the design of a neural network can be cumbersome. Inaccuracies may lead to a manifold of problems in the application such as higher errors due to local optima, overfitting and ill-conditioning of the network, especially when the number of observations are small. In this paper we provide a method to overcome these difficulties by regulating the flexibility of the network and by rendering measures for validating the final network. In particular a method is proposed to equilibrate the number of hidden neurons and the value of the weight decay parameter based on 5 and 10-fold cross-validation. In the validation process the performance of the neural network is compared with a linear model with the same input variables. The degree of non-linearity with respect to each explanatory variable is estimated by numerical differentiation. Furthermore we present a rationale quantifying the degree of confidence of the neural network predictions. The methods are illustrated in two case studies: predicting the price of housing in Boston metropolitan area and the classification of bond ratings.