Promoting Less Complex and More Honest Price Negotiations in the Online Used Car Market with Authenticated Data

Abstract

Online peer-to-peer (P2P) sales of used and or high-value goods are gaining more and more relevance today. However, since potential buyers cannot physically examine the product quality during online sales, information asymmetries and consequently uncertainty and mistrust that already exist in offline sales are exacerbated in online markets. Authenticated data platforms have been proposed to solve these problems by providing authenticated data about the negotiation object, integrating it into text-based channels secured by IT. Yet, we know little about the dynamics of online negotiations today and the impact of the introduction of authenticated data on online negotiation behaviors. We address this research gap based on two experimental studies along with the example of online used car trade. We analyze users’ communicative and strategic actions in current P2P chat-based negotiations and examine how the introduction of authenticated data affects these behaviors using a conceptional model derived from literature. Our results show that authenticated data can promote less complex negotiation processes and more honest communication behavior between buyers and sellers. Further, the results indicate that chats with the availability of authenticated data can positively impact markets with information asymmetries. These insights provide valuable contributions for academics interested in the dynamics of online negotiations and the effects of authenticated data in text-based online negotiations. In addition, providers of trade platforms who aim to advance their P2P sales platforms benefit by achieving a competitive advantage and a higher number of customers.

Hybrid forecasting model in used car price forecasting based on stochastic algorithm

“…The obtained results can be directly used as the input of the support vector machine SVM, eliminating the process of dimensionality reduction and finding features, reducing run time. Therefore, using the GBDT model, the new combined features trained are merged with the original variables, and the features of the original data are optimized to predict the price of used cars [11].…”