Vijay S. Desai

A comparison of neural networks and linear scoring models in the credit union environment

Abstract The purpose of the present paper is to explore the ability of neural networks such as multilayer perceptrons and modular neural networks, and traditional techniques such as linear discriminant analysis and logistic regression, in building credit scoring models in the credit union environment. Also, since funding and small sample size often preclude the use of customized credit scoring models at small credit unions, we investigate the performance of generic models and compare them with customized models. Our results indicate that customized neural networks offer a very promising avenue if the measure of performance is percentage of bad loans correctly classified. However, if the measure of performance is percentage of good and bad loans correctly classified, logistic regression models are comparable to the neural networks approach. The performance of generic models was not as good as the customized models, particularly when it came to correctly classifying bad loans. Although we found significant differences in the results for the three credit unions, our modular neural network could not accommodate these differences, indicating that more innovative architectures might be necessary for building effective generic models.

Interactions between members of a marketing-production channel under seasonal demand

Abstract Key decisions in a marketing-production channel faced with a seasonal demand include production and pricing decisions for the manufacturer, and processing and pricing decisions for the retailer. In such situations, a variety of contracts are possible, including contracts under which the manufacturer charges a constant price throughout the season, or the retailer processes at a constant rate throughout the season. Also possible are channel structures in which the manufacturer and retailer cooperate to make decisions jointly. The main objective of the present paper is to compare optimal policies under all the situations described above. We show that having a constant manufacturers price, or a constant processing rate does not have a significant impact upon the retailers price or inventory decisions. However, if the demand is not highly seasonal, a constant processing rate contract will result in higher production and processing rates, and a lower manufacturers price, compared to a constant manufacturers price contract. Also, a centralized channel involving joint decision making will lead to a lower price at the retail level, and higher production and processing rates. Hence, issues such as channel coordination and contracts between manufacturers and retailers, which have received significant attention in the marketing literature, have implications for the production area as well.

Marketing-production decisions under independent and integrated channel structure

The topic of channel structure has recently attracted much attention among researchers in the marketing and economics area. However, in a majority of the existing literature the cost considerations are extremely simplified with the major focus being pricing policy. What happens when cost incurring decisions are strongly connected with pricing policies? This is the theme we wish to explore in the present paper. The non-trivial costs considered are production, inventory, and retailer effort rate, i.e. we seek to explore the marketing-production channel. We have used the methodology of differential games. The open-loop Stackelberg solution concept has been used to solve the manufacturer and retailers problem. The Pareto solution concept has been used to solve the problem of the vertically integrated firm. The production, pricing, and effort rate policies thus derived have been compared to obtain insights into the impact of channel structure on these policies. Also, to examine the relation between channel structure and the retailing operation requiring effort, we derive the Stackelberg and Pareto solutions with and without effort rate as a decision variable. We show that once the production rate becomes positive, it does not become zero again. This implies production smoothing. However, none of the gains of production smoothing are passed on to the retailer. The optimal production rate and the inventory policy are a linear combination of the nominal demand rate, the peak demand factor, the salvage value, and the initial inventory. Also, as opposed to some of the existing literature, the optimal policies need not necessarily be concave in nature. In the scenario where the relating operation does not require effort, the pricing policies of the manufacturer and the retailer, and the production policy of the manufacturer have a synergistic effect. However, in the scenario where the retailing operation does benefit from effort, the retailers pricing policy need not necessarily be synergistic with other policies. With regard to channel structures, it seems that production smoothing will be done more efficiently in the integrated setup. Also, we show that the price paid by the consumer need not necessarily be lower in the integrated setup. But despite higher prices, the channel profits are higher in the integrated setup. This implies a conflict between the interests of the consumers and the firm. Also, this contradicts the results of some of the earlier papers that have used simple static models.

The relationship between R&D effort rate and the pricing of an exhaustible resource

We attempt a study of the optimal pricing strategy of a cartel selling an exhaustible resource and the optimal RD if the terminal market share is small, the entrant will give up with minimal resistance from the cartel; and if the terminal market share is medium, the cartel might allow entry. It is most likely that the cartels price after entry will decrease with time.