資產危機期間的多角化投資策略: 不動產投資信託與股票的尾端相依性

Diversification Strategies during Asset Crises: REIT-Stock Tail Dependence

劉世民(Shih-Min Liu)

指導教授 ： 吳志強

元智大學/管理學院/商學碩士班(財務金融學程)/碩士(2012年)

https://doi.org/10.6838/YZU.2012.00217

摘要

本文藉由不動產投資信託(REIT)與股票的價格報酬動態相依性，來探討多角化投資與資產配置策略，以聚焦於最近發生房市泡沫化與金融危機的這十年的資產市場。我們使用range-based波動模型，透過copula函數，來捕捉資產價格的波動度與不同資產間相依性的結構。我們發現厭惡風險的投資人，若採用range-based (asymmetric CARR)架構之動態投資組合時，所願意支付的轉換費用，高於採用return-based (GJR-GARCH)架構，基於此而證實range-based架構優於return-based架構。在最適資產配置的條件下，經濟價值的預測結果顯示，左尾的資產價格報酬相依性在2008-2009年大幅度提高，且動態投資策略具有顯著的經濟價值。此外，在資產市場危機時期，尋求能夠解釋跨資產相依性的經濟變數之困難度頗高。我們發現只有抵押貸款利差，可以提供額外的訊息，來幫助我們預測跨資產的右尾的相依性。

關鍵字

尾端相依性 ； 不動產投資信託 ； 資產配置 ； 動態copula ； 不對稱CARR 模型 ； GJR- GARCH模型

並列摘要

The study sheds insight into time-varying cross-asset linkages and asset-allocation strategies by investigating co-movements between REIT (real estate securitized trust) and stock price returns, spotlighting the recent decade with remarkable real estate cycles and asset crises. We propose a range-based volatility model with a dynamic copula function in order to characterize asset price volatility and cross-asset dependence structures. We further conduct out-of sample forecasts of the economic values based on the asset-allocation problem, and find a sustainable swing in REIT-stock lower tail dependence during 2008-2009. The findings suggest that the range-based volatility (asymmetric CARR) model is superior to the return-based (GJR- GARCH) framework since risk-averse investors are willing to pay more fees to switch from the dynamic portfolios based on the latter to those based on the former. We provide evidence of positive economic values of the dynamic strategies, and indicate the challenge of significantly explaining cross-asset linkages in asset busts: only the mortgage spread is informative in explaining REIT-stock upper tail dependence from a forward-looking perspective.

並列關鍵字

copula ； range-based volatility (asymmetric CARR) model ； return-based (GJR- GARCH) framework ； Tail dependence ； REIT (real estate securitized trust) ； real estate cycle

參考文獻

Alizadeh S, Brandt MW, Diebold FX. 2002. Range-based estimation of stochastic volatility models. The Journal of Finance 57(3): 1047-1091

Andersen TG, Bollerslev T. 1998. Answering the skeptics: yes, standard volatility models do provide accurate forecasts. International Economic Review 39(4): 885-905

Bollerslev T. 1986. Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics 31(3): 307-327

Brownlees CT, Gallo GM. 2010. Comparison of volatility measures: a risk management perspective. Journal of financial econometrics 8(1): 29-56.

Bu R, Giet L, Hadri K, Lubrano M. 2011. Modeling multivariate interest rates using time-varying copulas and reducible nonlinear stochastic differential equations. Journal of Finance and Economics 9(1),198-236