Dokument: Ideal Agent System with Triplet States: Model Parameter Identification of Agent–Field Interaction

Titel:Ideal Agent System with Triplet States: Model Parameter Identification of Agent–Field Interaction
URL für Lesezeichen:https://docserv.uni-duesseldorf.de/servlets/DocumentServlet?id=67295
URN (NBN):urn:nbn:de:hbz:061-20241031-100449-6
Kollektion:Publikationen
Sprache:Englisch
Dokumententyp:Wissenschaftliche Texte » Artikel, Aufsatz
Medientyp:Text
Autoren: Börner, Christoph J. [Autor]
Hoffmann, Ingo [Autor]
Stiebel, John H. [Autor]
Dateien:
[Dateien anzeigen]Adobe PDF
[Details]688,8 KB in einer Datei
[ZIP-Datei erzeugen]
Dateien vom 31.10.2024 / geändert 31.10.2024
Stichwörter:partition function, utility function, entropy, agent system, canonical ensemble, risk assessment
Beschreibung:On the capital market, price movements of stock corporations can be observed independent of overall market developments as a result of company-specific news, which suggests the occurrence of a sudden risk event. In recent years, numerous concepts from statistical physics have been transferred to econometrics to model these effects and other issues, e.g., in socioeconomics. Like other studies, we extend the approaches based on the “buy” and “sell” positions of agents (investors’ stance) with a third “hold” position. We develop the corresponding theory within the framework of the microcanonical and canonical ensembles for an ideal agent system and apply it to a capital market example. We thereby design a procedure to estimate the required model parameters from time series on the capital market. The aim is the appropriate modeling and the one-step-ahead assessment of the effect of a sudden risk event. From a one-step-ahead performance comparison with selected benchmark approaches, we infer that the model is well-specified and the model parameters are well determined.
Rechtliche Vermerke:Originalveröffentlichung:
Börner, C. J., Hoffmann, I., & Stiebel, J. H. (2023). Ideal Agent System with Triplet States: Model Parameter Identification of Agent–Field Interaction. Entropy, 25(12), Article 1666. https://doi.org/10.3390/e25121666
Lizenz:Creative Commons Lizenzvertrag
Dieses Werk ist lizenziert unter einer Creative Commons Namensnennung 4.0 International Lizenz
Fachbereich / Einrichtung:Wirtschaftswissenschaftliche Fakultät
Dokument erstellt am:31.10.2024
Dateien geändert am:31.10.2024
english
Benutzer
Status: Gast
Aktionen