How market signals shape online auction prices for remanufactured electronics: an explainable artificial intelligence approach
Yeunsoo Park et al.
Abstract
Purpose This study examines how prices are formed in online auctions for remanufactured electronics, focusing on the joint roles of sellers, buyers and listing signals in shaping final transaction outcomes. It aims to move beyond linear pricing assumptions by uncovering how these signals interact under different competitive conditions. Design/methodology/approach The study analyses 1,039 completed eBay UK auctions using an explainable machine-learning framework. A predictive model is estimated and interpreted using SHAPley Additive exPlanations to assess the nonlinear and interaction effects of pricing strategies, seller reputation, bidding activity and listing characteristics on final auction prices. Findings The results indicate that auction price formation is driven by nonlinear, context-dependent interactions rather than by stable marginal effects. Starting prices, seller credibility indicators, bidding intensity and listing presentation function as interdependent signals whose influence varies with the level of competition. Their effects vary with the surrounding auction environment, highlighting the conditional nature of value formation in remanufactured product markets. Originality/value This study contributes to the literature on online auctions and remanufactured product markets by demonstrating the usefulness of explainable machine learning for uncovering structured yet non-linear pricing mechanisms. It provides actionable insights into how sellers can align pricing strategies, credibility cues and information disclosure to support more effective price discovery, while avoiding restrictive linear modelling assumptions commonly used in prior research.
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.50 × 0.4 = 0.20 |
| M · momentum | 0.50 × 0.15 = 0.07 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.