人身保險公司商品組合最佳化

Life Insurance Port folio Optimization

大多數壽險公司在其業務組合中並未實施最佳化，且這方面的學術研究相對有限。本研究提出利用每個產品線的準備金百分比變化的變異數來最佳化壽險公司的業務組合。我們運用了隨機矩陣理論（RMT）以及馬可維茲投資組合理論（MPT）來構建最佳業務組合。利用2001年至2021年的美國保險監理官協會（NAIC）數據，我們證明了這兩種方法得到的準備金百分比變化的變異數均小於歷史變異數。此外，相較於MPT，RMT在股票型保險公司方面表現優異，在相互保險公司方面與MPT相當。因此，壽險公司應在其業務組合中實施最佳化。

Most life insurance companies do not implement optimization in their business portfolios, and only limited academic research exists on this subject. This study proposes using the variances of the percentage changes in reserves for each product line to optimize the business portfolio for a life insurer. We apply the Random Matrix Theory (RMT) as well as the Markowitz Portfolio Theory (MPT) to construct the optimal portfolio. Using the U.S. National Association of Insurance Commissioners (NAIC) data from 2001 to 2021, we demonstrate that the variances of the percentage changes in reserves obtained through these two approaches are smaller than the historical variances. Furthermore, RMT performs superior to MPT in regard to stock insurers and equivalent to MPT regarding mutual insurers. Life insurers therefore would be well-advised to implement optimization in their business portfolios.