|
||||
|
||||
A Competing Risk Dynamic Hazard Approach to Investigate the Impairment Outcomes of Property-Casualty InsurersHuong D. DangUniversity of Canterbury - Economics and Finance November 17, 2010 Finance and Corporate Governance Conference 2011 Paper Abstract: This study applies a competing risk approach and develops dynamic type-specific Cox’s hazard models to investigate the underlying risk factors associated with the incidence and the timing of impairment events occurring to property-casualty insurers in the U.S. during January 1998 - June 2010. The models treat impairment events of different manners as distinct, and accounts for the changes in the financial characteristics of insurers as well as the development of macro-economic conditions over time. It is found that the hazards of impairment events of different manners are neither equal nor proportionate. The hazards for a supervision, a conservation, and a rehabilitation event increase more rapidly with time than the hazard for a liquidation event. The model for a generic impairment event and the models for manner-specific impairment events feature different significant underlying risk factors. The differences across estimated models are not due to random variations, which strengthen the evidence that impairment events of different manners are distinct phenomenon, and emphasize the need to develop separate models for separate impairment event types. The validation results over a five-year forecast horizon show that the models for manner-specific impairment events exhibit better predictive accuracy in forming time-varying survival estimates than does the model for a generic event. The proposed approach aids insurance regulators in the early identification of troubled insurers, in the choice of optimal means of regulatory interventions, and in the preparation of appropriate preventative/remedial plans with sufficient lead time.
Number of Pages in PDF File: 46 Keywords: Property-casualty insurers, financial impairment, hazard model, time-varying covariates, forecast accuracy JEL Classification: C14, C34, C41, C52, G33 working papers seriesDate posted: December 3, 2010 ; Last revised: January 16, 2012Suggested CitationContact Information
|
|
||||||||||||
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
FAQ
Terms of Use
Privacy Policy
Copyright
This page was processed by apollo7 in 0.360 seconds