Learning Negotiation Support Systems in Competitive Negotiations : A Study of Negotiation Behaviors and System Impacts

Prior research on negotiation support systems (NSS) has paid limited attention to the information content in the observed bid sequences of negotiators. Cognitive characteristics of individual negotiators and their impacts on negotiation performance have also received limited attention. In this paper, we examine the accuracy of an exponential functional form in representing observed bid sequences. Results indicate that the exponential function is a good model for observed bids. To leverage the information content in observed bids, we predict reservation values of negotiators based on their observed bids. Based on the negotiation behaviors of human subjects, we find that accurate estimates of opponent reservation values lead to improved negotiation outcomes. Results also show that bids are correlated with reservation values. We study the impact of negotiation support systems in helping users realize superior negotiation outcomes. Results indicate that automated negotiations using NSS can lead to superior negotiation outcomes compared to human subjects