Application of Neural Networks to House Pricing and Bond Rating

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 is 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 monotonicity with respect to each explanatory variable is calculated by numerical differentiation. The outcomes of this analysis is compared to what is expected from economic theory. Furthermore we propose a scheme for the application of monotonic neural networks to problems where monotonicity with respect to the explanatory variables is known a priori. The methods are illustrated in two case studies: predicting the price of housing in Boston metropolitan area and the classification of bond ratings.