Insolvency Prediction in Insurance Companies Using Support Vector Machines and Fuzzy Kernel C-Means

Abstract

Insolvency of insurance companies has been a concern of parties such as insurance regulators, investors, management, financial analysts, banks, auditors, policy holders, and consumers. This concern has arisen from the perceived need to protect the general public against the consequences of insurers insolvencies, as well as minimizing the responsibilities for management and auditors. In this paper we propose an approach to avoid insolvency in insurance companies. A large number of methods such as discriminant analysis, logit analysis, recursive partitioning algorithm, etc., have been used in the past for insolvency prediction. However, the special characteristics of the insurance sector have made most of them unfeasible, and just a few have been applied to this sector. In this study we predict the insolvency using two different methods, there are Support Vector Machines (SVM) and Fuzzy Kernel C-Means (FKCM). The results are very encouraging and show that SVM and FKCM can be a useful tool for parties who are interest in evaluating insolvency of an insurance firm.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 38 citations exist upstream.