A study on solvency of Life Insurance Industry in Taiwan

• Main Authors: Jiun-Hung Wu, 吳俊鴻
• Other Authors: Ming-Chun Lin
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/p4ar33

碩士 === 國立高雄第一科技大學 === 風險管理與保險所 === 96 === This research applies Discriminant analysis, Factor analysis, and Score-method to establish the forewarning system of solvency and insolvency for the life insurance companies in Taiwan from 2004 to 2006. Owing to insufficient data of the inland insolvent companies, this research use the estimated RBC ratio first and fit with the concept of normal distribution to obtain the degree of evaluation to discriminate the life insurance companies that have good solvency and life insurance companies that have bad solvency. On the other hand, apply factor analysis to reduce the variables into five factors, which are defined as “ benefit ability index”, “ management ability index ”,“ financial structure index ”, “ non-performing loans index ”, and “ debt paying ability index ”. Finally, apply the linear discriminant analysis and logistic regression analysis, then combine with all variables and stepwise discriminant two types, and combine with the “ 1:1 ”, “ 2:1 ” and “ the whole data ” three types of samples, altogether establish twelve kinds of models, to find out the forewarn models of solvency of life insurance industry. We find out that a model “linear discriminant analysis combine with all variables and samples in 1:1 type”, has better discriminant ability. For the life insurance companies that have good solvency, the average accuracy rate reaches 70.8%, and for the life insurance companies that have bad solvency, the average accuracy rate reaches 95.8%, and in all companies, the average accuracy rate reaches 83.3% .