下方風險對報酬績效之影響分析：以美國社會責任投資及反社會投資為例

NHUIR > College of Management > Department of Finance, The M.A. Program of Financial Management > Disserations and Theses > Item 987654321/19937

Please use this identifier to cite or link to this item: http://nhuir.nhu.edu.tw/handle/987654321/19937

Title:	下方風險對報酬績效之影響分析：以美國社會責任投資及反社會投資為例
Other Titles:	Disentangling the Effect of Downside Risk on Performance: Evidence for Socially Responsible and Anti-social Investing in the United States
Authors:	陳易佐 Chen, Yi-tso
Contributors:	財務金融學系財務管理碩士班張簡彰程;張瑞真 Chang-cheng Chang-chien;Jui-chen Chang
Keywords:	風險值;社會責任投資;Gram-Charlier展開式 Gram-Charlier Expansions;Value-at-Risk;Socially Responsible Investing
Date:	2009
Issue Date:	2015-03-24 16:53:48 (UTC+8)
Abstract:	社會責任投資(Socially Responsible Investing, SRI)的概念在過去十年來快速地形成，過去的學術文獻主要集中在探討社會責任投資的發展以及績效表現，然而與其對應的反社會投資(Anti-Social Investing, ASI)之研究則相對較少。本文以風險值(Value-at-Risk)的角度進一步衡量這兩種投資策略的下方風險，樣本資料包括十支美國社會責任投資基金、一支美國反社會投資基金以及兩個相關指數的日報酬資料，研究期間為2002年9月16日至2007年9月16日。除了使用傳統資本資產定價模型的績效衡量方法，本研究使用風險值修正夏普指數以及風險值調整後超額報酬來衡量樣本投資組合的績效表現。實證結果顯示，社會責任投資樣本組相較於反社會投資樣本組普遍具有較高的下方風險以及較低的年化平均報酬，而資料型態的差異(基金或指數)會影響其績效和下方風險的推論。此外， Wilcoxon符號等級檢定結果顯示風險值修正後之績效與修正前之績效存在顯著差異。在風險值調整後報酬之架構下，以國際證券交易所編制的SINdex指數之績效表現最佳。在風險值模型的實證結果部份，根據Kupiec (1995)非條件涵蓋比率檢定以及平均失敗誤差之統計結果，經Gram-Charlier展開式調整的風險值模型在各種風險值估計方法中具有最佳的準確性並保有效率性，此方法同時亦可解決常態分配尾部風險值低估問題而對金融資產報酬具有較佳的厚尾捕捉能力。　　The concept of socially responsible investing (SRI) has escalated rapidly over the past decade. Considerable interest in the evolution and performance of socially responsible investing continues to be widened in academic literature while the studies pertaining to its antithetical counterpart, anti-social investing (ASI), are relatively underappreciated. This article sheds further light on the downside risk of the two antipodal investment strategies, introducing a more robust measure of risk in the form of Value-at-Risk (VaR). The data are sourced from daily closing prices of ten U.S. SRI mutual funds, one U.S. ASI mutual fund and two relevant indices. The sample spans the period from September 16th, 2002 to September 16th, 2007. Apart from the traditional performance measurement of capital asset pricing model (CAPM), the VaR-modified Sharpe index and excess VaR-adjusted return are employed to conduct performance assessment for the correspondingly screened portfolios. The empirical evidence indicates that generally the SRI sample group embraces higher permeation of downside risks and lower annualized returns compared with the ASI sample group, whereas different data type (mutual fund or index) leads to variant downside risk and performance inferences. Furthermore, the result of the Wilcoxon signed-rank test indicates that the VaR-modified performance differs from the non-modified one significantly. Under the framework of the VaR-adjusted return the SINdex constructed by International Securities Exchange outperforms the rest of the reference portfolios. The empirical result from VaR modelling suggests that the VaR model adjusted by Gram-Charlier expansions is the most accurate measurement and maintains model efficiency among various estimation models in terms of Kupiec's (1995) unconditional coverage test and average failure bias. Simultaneously the model can address the VaR underestimation problem beneath the normality assumption, possessing superior competence for capturing heavy-tailed idiosyncrasy of financial asset returns.
Appears in Collections:	[Department of Finance, The M.A. Program of Financial Management] Disserations and Theses

Files in This Item:

File	Description	Size	Format
097NHU05304017-001.pdf		940Kb	Adobe PDF	373	View/Open
index.html		0Kb	HTML	361	View/Open

Loading...