
TL;DR
This paper develops a model of predictive scoring where an intermediary aggregates features to predict sender quality, balancing feature weighting to deter distortion and improve decision accuracy.
Contribution
It characterizes an optimal scoring rule that underweights and overweights features to maximize accuracy while deterring sender manipulation.
Findings
Optimal scoring rule underweights some features to prevent distortion.
Overweights certain features to ensure the score is accurate on average.
The proposed scoring rule outperforms full disclosure in mitigating commitment issues.
Abstract
I introduce a model of predictive scoring. A receiver wants to predict a sender's quality. An intermediary observes multiple features of the sender and aggregates them into a score. Based on the score, the receiver makes a decision. The sender prefers "higher" decisions, and she can distort each feature at a privately known cost. I characterize the scoring rule that maximizes decision accuracy. This rule underweights some features to deter sender distortion, and overweights other features so that the score is correct on average. The receiver prefers this scoring rule to full disclosure because it mitigates his commitment problem.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
