Comparing the predictive performance of a neural network model with some traditional market response models

International Journal of ForecastingPublished 1 September 1994

Chanda Ghose Dasgupta, Gary S. Dispensa, Sanjoy Ghose

DOI10.1016/0169-2070(94)90004-3

Source URL

Citations124

SJR quartileQ1

SJR score2.43

SNIP3.36

TL;DR

This study believes that its study is one of the first applications of neural network models using individual level cross-sectional survey data for predicting market response and that its findings have opened the doors to further research about the applicability of Neural Network Modeling using Individual levelCross-sectional data, as opposed to using aggregate company level data.

Abstract

The study compares the performance of two statistical market response models (a logistic regression model and a discriminant analysis model) to that of a back propagation neural network model. The comparative performances of these models are evaluated with respect to their ability to identify consumer segments based upon their willingness to take financial risks and to purchase a non-traditional investment product. The empirical analysis is conducted using two different real-world individual level cross-sectional data sets related to the marketing of financial services. If we rank order the performance of these models, we find that the neural network model performs better than the other two models. However, the level of performance is not significantly higher than those of the other models. This is in contradiction to the findings in the financial industry applications in the literature, where neural network models have, in general, significantly outperformed traditional statistical response models. We believe that our study is one of the first applications of neural network models using individual level cross-sectional survey data for predicting market response and that our findings have opened the doors to further research about the applicability of neural network modelling using individual level cross-sectional data, as opposed to using aggregate company level data.

Keywords

Computer ScienceDecision SciencesBusiness, Management and Accounting

NatureLearning representations by back-propagating errors

30,885 Citations1986David E. Rumelhart, Geoffrey E. Hinton +1 more

Back-propagation repeatedly adjusts the weights of the connections in the network so as to minimize a measure of the difference between the actual output vector of the net and the desired output vector, which helps to represent important features of the task domain.

Proceedings of the National Academy of SciencesNeural networks and physical systems with emergent collective computational abilities.

19,349 Citations1982J. J. Hopfield

A model of a system having a large number of simple equivalent components, based on aspects of neurobiology but readily adapted to integrated circuits, produces a content-addressable memory which correctly yields an entire memory from any subpart of sufficient size.

Bulletin of Mathematical BiologyA logical calculus of the ideas immanent in nervous activity

17,984 Citations1943Warren S. McCulloch, Walter Pitts

Management ScienceManagerial Applications of Neural Networks: The Case of Bank Failure Predictions

1,124 Citations1992Kar Yan Tam, Melody Y. Kiang

Empirical results show that neural nets is a promising method of evaluating bank conditions in terms of predictive accuracy, adaptability, and robustness.

Decision Support SystemsBankruptcy prediction using neural networks

721 Citations1994Rick L. Wilson, Ramesh Sharda

The study indicates that neural networks perform significantly better than discriminant analysis at predicting firm bankruptcies, and implications for the accounting professional, neural networks researcher and decision support system builders are highlighted.

Decision SciencesNeural Networks: A New Tool for Predicting Thrift Failures*

534 Citations1992Linda Salchenberger, E. Mine Çinar +1 more

A neural network model that processes input data consisting of financial ratios is developed to predict the financial health of thrift institutions and its ability to discriminate between healthy and failed institutions is compared to a traditional statistical model.

SIMULATIONTime series forecasting using neural networks vs. Box- Jenkins methodology

406 Citations1991Zaiyong Tang, Chrys de Almeida +1 more

It is found that for time series of different complexities there are optimal neural network topologies and parameters that enable them to learn more efficiently and are also parsimonious in their data requirements.

Physical TherapyNonparametric Methods for Quantitative Analysis

295 Citations1977Wayne W. Daniel

This is the book that many people in the world waiting for to publish, nonparametric methods for quantitative analysis, and the book lovers are really curious to see how this book is actually.

IEEE International Conference on Neural NetworksBond rating: a nonconservative application of neural networks

216 Citations1988Dutta, Shekhar

The authors apply neural networks to a generalization problem of predicting the ratings of corporate bonds, where conventional mathematical modeling techniques have yielded poor results and it is difficult to build rule-based artificial-intelligence systems.

Neural networks for bond rating improved by multiple hidden layers

133 Citations1990Alvin J. Surkan, J. Clay Singleton

These experiments suggest that networks trained with two hidden layers outperform a network having only one hidden layer containing a comparable number of processing elements, when the task was presented as one of separating two noncontiguous classes.