Estimation of Premium Rates With Scarce Data: A Simulation Study on Aquaculture Insurance

The general purpose of this thesis is to investigate the small sample performances of various rating methodologies in estimating premium rates for catfish insurance policy. To accomplish this objective, we simulate the yield data by modeling the frequency, severity, temporal and spatial correlations of twenty major risk factors in catfish production. In order to increase the applicability of this simulation study, twelve scenarios are considered to see how the methodologies will perform under different assumptions. Simulation results show that, under these data generating processes, estimators that use extraneous data generally perform better than those that only use individual data when the sample size is small.

zhihua_li_masters_thesis.pdf (593.1 KB)

Author(s)

Publication Date