本研究欲探討IFRS 9 (International Financial Reporting Standards 9) 與IFRS 17同步實施後，保險合約負債與多數資產須透過公允價值衡量時，壽險公司應如何對其終身壽險負債之組成進行妥善資產配置，以降低損益和權益變動，同時維持獲利。
本研究以GARCH模型模擬股價變化，以Svensson (1994)利率模型之隱含參數配適VAR (Sims, 1986)模型並模擬利率走勢，以終身壽險為負債組成，藉此模擬未來三年資產負債變化，並檢視第三年末資產負債分布情形，以資產報酬率及其標準差、業主權益標準差及破產機率為指標，比較不同權重產生各個指標的數值，最後將各個面向納入目標函數，以了解不同目標下權重將如何變化。
結果顯示全投股票破產機率最高，二成股票和八成債券可組成最高資產報酬率；決策的考量若包含權益標準差，將驅使權重分配至債券部位，因債券與負債的變化皆和利率有關，能夠互相抵銷；決策的考量包含資產報酬率標準差，將使權重移動至其他綜合損益或透過攤銷後成本衡量，以上的動態關係，可作為壽險公司資產配置決策的參考。

Once IFRS 9 and IFRS 17 are officially launched simultaneously, the insurance contract liabilities and most financial assets should be measured at fair value. The objective of this article is to analyze how life insurers do the asset allocation that aims to get high returns and lower deviation of profit and loss and equity in response to the whole-life insurance contracts under IFRS 9 and IFRS 17.
This article based on the GARCH model to simulate the path of stock price. In addition, we use the parameters implied by Svensson interest rate model to fit VAR model in order to simulate the path of interest rate. Furthermore, the liability comprises whole life insurance. We simulate the path of asset and liability for 3 years and then focus on the distribution of asset and liability in 3rd year. We compare the values from different weight set like returns on assets, standard deviation of returns on assets, standard deviation of equity and default risk. Moreover, we design the object function that can help us understand how the weight will change given different goals.
One of the result shows that investing all assets on stocks creates the highest probability of default risk. The combination of 20% stock and 80% bond creates the highest return on assets. Besides, if one of the component in the object function is standard deviation of equity, it will drive weight to bond investment so that asset can offset part of liabilities because the change of bond value and whole-life insurance both connected to interest rate. If the component includes standard deviation of returns on assets, it will drive weight to other comprehensive income or amortization cost. All of these are the dynamic effects that can help life insurers for decision of asset allocation.

中文部分
金融監督管理委員會國際財務報導準則下載專區，上網日期106年11月28日，檢自：http://163.29.17.154/ifrs/index.cfm?act=ifrs_2016
財團法人中華民國櫃檯買賣中心債券市場資訊殖利率曲線與技術手冊
高渭川、周寶蓮，2015。《國際財務報導準則第四號(IFRS 4)-保險合約會計第二階段研究案》（金管會委託研究計畫 10403-0008）。台北：金融監督管理委員會保險局。
英文部分
Bollerslev, T., 1986, Generalized autoregressive conditional heteroskedasticity. Journal of econometrics, 31 (3) 307-327.
Pooter, M. D., 2007, Examining the Nelson-Siegel class of term structure models (No. 07-043/4). Tinbergen Institute Discussion Paper.
Sims, C. A., 1986, Are forecasting models usable for policy analysis？. Quarterly Review, (Win), 2-16.
Svensson, L. E., 1994, Estimating and interpreting forward interest rates: Sweden 1992-1994 (No. w4871). National Bureau of Economic Research.
ASAF Agenda Paper 11, The overlay approach, Retrieved May 28 2017, from:http://www.ifrs.org/Meetings/MeetingDocs/ASAF/2015/December/1512-ASAF-11-The-overlay-approach.pdf
An option for presenting the effect of changes in discount rates, Retrieved May 28 2017, from:http://www.ifrs.org/Meetings/MeetingDocs/IASB/2014/March/02E%20IC%20OCI%20paper.pdf