Document Type

Article

Department

Institute for Educational Development, East Africa

Abstract

The incidence of metastatic cancer over the years has gradually increased. This can be due to delayed quality treatment, attributed to insufficient funds. Models that can more accurately price cancer insurance policies incorporating its metastatic nature would enable insurance firms to develop affordable and profitable cancer insurance policies. Effectively marketed and reasonably priced insurance policies can enhance the uptake of cancer specific insurance, thereby alleviating the financial strain linked to cancer treatment. To capture the impact of metastatic cancer on overall cancerrelated losses, we construct Katz recursive models grounded in the Katz family of distributions, specifically the (a, b, 0) class, using Kenyan data. These distributions will consider both the claim count and the claim amount of metastatic cancer. The Katz family (a, b, 0) distribution consists of four distributions, which will be applied in their recursive form. Katz distributions offer a robust framework for analyzing nested count effects in the estimation of total losses. The Katz-Poisson distribution closely mirrored the actual count probabilities, particularly within the Katz family of (a, b, 0) distributions, where it demonstrated the best fit. Unlike the conventional Poisson model, the Katz-Poisson approach enhances count estimation by incorporating the transition probabilities associated with metastatic cancer progression. Regarding claim amount, the probabilities derived from the Generalized Pareto distribution showed an upward trend when comparing the three-phase and six-phase models, aligning with observed data patterns. Overall, the Katz-Poisson combined with the Generalized Pareto distribution yielded the most accurate representation for modeling within the Katz (a, b, 0) family.

Publication (Name of Journal)

The Egyptian Statistical Journal (ESJ)

DOI

doi: 10.21608/esju.2025.395066.1100

Creative Commons License

Creative Commons Attribution-Share Alike 4.0 International License
This work is licensed under a Creative Commons Attribution-Share Alike 4.0 International License.

Share

COinS