Please use this identifier to cite or link to this item: https://ah.lib.nccu.edu.tw/handle/140.119/47219
題名: Valuation and Hedging of Limited Price Indexation (LPI) Liability
作者: 黃泓智
Huang,H.-C.;Cairns,A.J.G
關鍵詞: LPI Liability; Static Hedging; Regular Rebalancing; Dynamic Hedging; Risk Minimising Hedging
日期: 2004
上傳時間: 17-Oct-2010
摘要: This paper considers the market or economic valuation and the hedging of Limited Price Indexed (LPI) liabilities. This involves finding optimal static and dynamic hedging strategies which minimise the riskiness of the investment portfolio relative to the liability.\r\nIn this paper we do not aim to find the perfect hedge in a perfect world. Instead, it is assumed that optimisation is restricted to three commonly used asset classes in pension funds: cash; long-term (or irredeemable) fixed-interest bonds; and long-dated index-linked bonds. The economic value of the liability is then defined as the value of the best matching portfolio using a mean/variance type of loss function. Specifically, we adopt the risk minimising approach of Föllmer & Sondermann (1986) and Schweizer & Föllmer (1988). Even with such a simple loss function, establishing the theoretically optimal solution can be difficult. We propose that a practical solution close to the theoretical optimum can be found using two approximations. First, we approximate the ‘true’ stochastic economic model by a vector autoregressive model of order one. Second, we use a sequence of linearisations to approximate non-linear by straightforward quadratic minimisation problems.\r\nThe proposed approach is illustrated with various numerical examples, and we compare the results of the approximately optimal hedging strategy with static strategies.
關聯: British Actuarial Journal,10(3), 627-663
資料類型: article
DOI: http://dx.doi.org/10.1017/S1357321700002713
Appears in Collections:期刊論文

Files in This Item:
File Description SizeFormat
627663.pdf392.2 kBAdobe PDF2View/Open
Show full item record

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.