Titel: Using copulas for rating weather index insurance contracts
Autoren: Bokusheva, Raushan
Erschienen in: Journal of Applied Statistics
Verlag / Hrsg. Institution: Routledge
Erscheinungsdatum: 29-Jan-2018
Sprache: Englisch / English
Schlagwörter: Extreme dependence modeling; Copula; Index-based insurance; Agriculture; Risk
Fachgebiet (DDC): 360: Soziale Probleme, Sozialdienste, Sozialversicherungen
630: Landwirtschaft
Zusammenfassung: This study develops a methodology for a copula-based weather index insurance design. Because the copula approach is better suited for modeling tail dependence than the standard linear correlation approach, its use may increase the effectiveness of weather insurance contracts designed to provide protection against extreme weather events. In our study, we employ three selected Archimedean copulas to capture the left-tail dependence in the joint distribution of the farm yield and a specific weather index. A hierarchical Bayesian model is applied to obtain consistent estimates of tail dependence using relatively short time series. Our empirical results for 47 large grain-producing farms from Kazakhstan indicate that, given the choice of an appropriate weather index to signal catastrophic events, such as a severe drought, copula-based weather insurance contracts may provide significantly higher risk reductions than regression-based indemnification schemes.
Departement: Life Sciences und Facility Management
Organisationseinheit: Institut für Umwelt und Natürliche Ressourcen (IUNR)
Publikationstyp: Beitrag in wissenschaftlicher Zeitschrift / Article in scientific Journal
Art der Begutachtung: Peer review (Publikation)
DOI: 10.1080/02664763.2017.1420146
ISSN: 1360-0532
URI: https://digitalcollection.zhaw.ch/handle/11475/3432
Lizenz (gemäss Verlagsvertrag): Lizenz gemäss Verlagsvertrag / Licence according to publishing contract
Enthalten in den Sammlungen:Publikationen Life Sciences und Facility Management

Dateien zu dieser Ressource:
Es gibt keine Dateien zu dieser Ressource.

Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.