Working Paper: NBER ID: w29127
Authors: Can Urgun; Leeat Yariv
Abstract: We study a model of retrospective search in which an agent—a researcher, an online shopper, or a politician—tracks the value of a product. Discoveries beget discoveries and their observations are correlated over time, which we model using a Brownian motion. The agent, a standard exponential discounter, decides the breadth and length of search. We fully characterize the optimal search policy. The optimal search scope is U-shaped, with the agent searching most ambitiously when approaching a breakthrough or when nearing search termination. A drawdown stopping boundary is optimal, where the agent ceases search whenever current observations fall a constant amount below the maximal achieved alternative. We also show special features that emerge from contracting with a retrospective searcher.
Keywords: No keywords provided
JEL Codes: C61; C73; D25; D83
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Agent's search scope (L85) | Distance between current discovery and best previous discovery (Y80) |
| Stopping boundary (F55) | Current observation falling below maximum achieved value (C22) |
| Optimal search policy (D83) | Agent's patience (L85) |
| Contract structure (L14) | Search efficiency (G14) |
| Search costs (G19) | Agent's decision-making process (D80) |
| Previous discoveries (Y80) | Agent's search decisions (L85) |