dorsal/arxiv
View SchemaPhase coexistence in a forecasting game
| Authors | Philippe Curty, Matteo Marsili |
|---|---|
| Categories | |
| ArXiv ID | physics/0506151 |
| URL | https://arxiv.org/abs/physics/0506151 |
| DOI | 10.1088/1742-5468/2006/03/P03013 |
Abstract
Individual choices are either based on personal experience or on information provided by peers. The latter case, causes individuals to conform to the majority in their neighborhood. Such herding behavior may be very efficient in aggregating disperse private information, thereby revealing the optimal choice. However if the majority relies on herding, this mechanism may dramatically fail to aggregate correctly the information, causing the majority adopting the wrong choice. We address these issues in a simple model of interacting agents who aim at giving a correct forecast of a public variable, either seeking private information or resorting to herding. As the fraction of herders increases, the model features a phase transition beyond which a state where most agents make the correct forecast coexists with one where most of them are wrong. Simple strategic considerations suggest that indeed such a system of agents self-organizes deep in the coexistence region. There, agents tend to agree much more among themselves than with what they aim at forecasting, as found in recent empirical studies.
{
"annotation_id": "340016cc-b233-402c-9ca1-73ad61d32d67",
"date_created": "2026-03-02T18:01:00.808000Z",
"date_modified": "2026-03-02T18:01:00.808000Z",
"file_hash": "92131580c389d65b3aa4c6261eb797401307dd1a23b6095e09266cceb6f0b3c3",
"private": false,
"record": {
"abstract": "Individual choices are either based on personal experience or on information\nprovided by peers. The latter case, causes individuals to conform to the\nmajority in their neighborhood. Such herding behavior may be very efficient in\naggregating disperse private information, thereby revealing the optimal choice.\nHowever if the majority relies on herding, this mechanism may dramatically fail\nto aggregate correctly the information, causing the majority adopting the wrong\nchoice. We address these issues in a simple model of interacting agents who aim\nat giving a correct forecast of a public variable, either seeking private\ninformation or resorting to herding. As the fraction of herders increases, the\nmodel features a phase transition beyond which a state where most agents make\nthe correct forecast coexists with one where most of them are wrong. Simple\nstrategic considerations suggest that indeed such a system of agents\nself-organizes deep in the coexistence region. There, agents tend to agree much\nmore among themselves than with what they aim at forecasting, as found in\nrecent empirical studies.",
"arxiv_id": "physics/0506151",
"authors": [
"Philippe Curty",
"Matteo Marsili"
],
"categories": [
"physics.soc-ph",
"cond-mat.dis-nn",
"nlin.AO",
"q-fin.TR"
],
"doi": "10.1088/1742-5468/2006/03/P03013",
"title": "Phase coexistence in a forecasting game",
"url": "https://arxiv.org/abs/physics/0506151"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "cfcde2a3-49f5-4052-ad14-a1b0ef2696bf",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}