Adverse Selection and Auction Design for Internet Display Advertising

Abstract

We model an online display advertising environment in which “performance” advertisers can measure the value of individual impressions, whereas “brand” advertisers cannot. If advertiser values for ad opportunities are positively correlated, second-price auctions for impressions can be inefficient and expose brand advertisers to adverse selection. Bayesian-optimal auctions have other drawbacks: they are complex, introduce incentives for false-name bidding, and do not resolve adverse selection. We introduce “modified second bid” auctions as the unique auctions that overcome these disadvantages. When advertiser match values are drawn independently from heavy-tailed distributions, a modified second bid auction captures at least 94.8 percent of the first-best expected value. In that setting and similar ones, the benefits of switching from an ordinary second-price auction to the modified second bid auction may be large, and the cost of defending against shill bidding and adverse selection may be low. (JEL D44, D82, L86, M37)

Wireless Edge-Empowered Metaverse: A Learning-Based Incentive Mechanism for Virtual Reality

“…The value of simulations for each simulation pair of AV i and virtual simulator k is U i,k , which is the product of the common value v i of AV i and the match quality m i,k , i.e., U i,k " v i m i,k . The common values for every virtual simulator k are gained from the provisioning of traffic simulation for the AV i, which can be represented by the AV i's private value v i [35]. Additionally, the amount of personalized information determines the match quality m i,k of virtual simulator k. This way, the values of AVs and virtual simulators in autonomous driving systems are positively correlated.…”

“…However, the driving virtual simulator 0 that provides driving simulation to the AV ι cannot immediately measure its match quality m ι,0 . Therefore, asymmetric information exists among virtual simulators that might result in adverse selection [35].…”

“…To tackle the multi-task DT offloading problem in autonomous driving systems with traffic and driving simulations empowered by generative AI, we propose the multitask enhanced second-score auction-based mechanism, named MTEPViSA, based on the EPViSA proposed in [20]. Integrating the multi-dimensional auction [42] and the enhanced second-price auction [35], the MTEPViSA consists of four components in the mechanism, including the bidding process, the scoring rule, the allocation rules, and the pricing rules.…”

Impression Effect vs. Click-through Effect: Mechanism Design of Online Advertising

“…Bidders in search advertising are always "performance advertisers" who place ads in order to elicit an immediate performance, whereas many display ads are placed by "brand advertisers," who are instead building awareness [3,4]. However, advertisers could achieve two effects simultaneously no matter which kind of advertising format is adopted.…”

Generalized Second Price Auctions over a Network

“…Recent work has focused on efficiency of GSP auctions (see Gomes and Sweeney [9]), inefficiency under uncertainty (see Caragiannis, et al [10]), as well as on seller's revenue (see Yenmez [11]). For a model where some advertisers have limited information regarding the value of the display ads, see Arnosti, Beck, and Milgrom [12]. Athey and Ellison [13] consider auction design questions when advertising consumers search optimally.…”