Matrix Variate Distributions as a Tool for Insurers and their Application to Natural Hazard Loss Modeling

Project: Research project

Project Details

Description

Matrix Variate Distributions as a Tool for Insurers and their Application to Natural Hazard Loss Modeling Matrix Variate Distributions as a Tool for Insurers and their Application to Natural Hazard Loss Modeling A growing number of insurance problems require modeling the distributions of random matrices. The most commonly used of such distributions within the nance and actuarial literature is the Wishart distribution [1], which arises as the distribution of the sample covariance matrix of a multivariate normal dataset. The Wishart distribution has widely been applied in insurance and nance problems such as portfolio analysis [2], time series analysis [3], and rate making [4]. However, many other matrix variate distributions exist, such as the matrix variate normal and matrix variate t distributions, and their use has not been explored yet within the actuarial domain.
StatusActive
Effective start/end date7/8/217/7/24

Funding

  • Casualty Actuarial Society (CAS): $17,000.00

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.