透過您的圖書館登入
IP:3.144.42.196
  • 學位論文

風險預測、條件風險值與最適投資組合績效

Risk Measure, Conditional VaR and the Performance of Portfolio Optimization

指導教授 : 許江河
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


由於金融資產報酬波動性是影響投資組合績效的關鍵因素之一,因此若能夠充份掌握投資標的波動性,便能建構績效較佳的投資組合,故本研究主要探討波動性預測對於資產配置績效的影響。本研究建構投資組合的標的資產是台灣50指數及台灣中型100指數之成分股,研究期間為2003年6月至2008年4月。樣本資料採用固定窗口 (fixed window)處理方式,風險預測模型則有等加權移動平均法、指數加權移動平均法與拔靴複製法三種。經由此三種方法預測投資組合的風險值 (Value-at-Risk, VaR)及條件風險值 (Conditional Value-at-Risk, CVaR),再透過規劃求解獲得條件風險值最小化的最適投資組合之後,即可進行投資組合績效比較。本研究歸納以下兩點實證結果: (一)比較不同風險預測模型建構而成的投資組合績效與臺灣加權股價指數之報酬率的績效,結果顯示三種風險預測模型所建構的投資組合績效皆較臺灣加權股價指數績效為佳。 (二)最適資產配置取決於所估計的風險值及條件風險值;亦即不同的風險模型會給予不同的樣本分配及參數,進而預測出不同的波動度,以影響資產配置績效表現。在這三種風險預測模型中,當信賴水準 為95%時,以拔靴複製法的績效表現最好;當信賴水準 為99%時,以等加移動平均法的績效表現最好。

並列摘要


Since the return volatility of financial assets plays an important role on the performance of portfolios, investors can improve the performance of their portfolios by controlling the volatility of their assets. Therefore, this study examines the influence of the risk estimation on the performance of portfolios. The data used in this study consists of daily returns of the 150 listed companies in the TSEC Taiwan 50 index and TSEC Taiwan Mid-Cap 100 Index and spans from June, 2003 to April, 2008. Under the framework of the fixed window approach, three risk measures, namely the equally weighted moving average model, the exponentially weighted moving average model, and the bootstrap simulation model, are employed to predict the Value-at-Risk and the Conditional Value-at-Risk of the portfolios. After solving the minimization problems of the Conditional Value-at-Risk of the portfolios, the optimal portfolios could be held and their performances could then be compared. The results of this study are shown as follows: (1) All of the optimal portfolios built by minimizing the Conditional Value-at-Risk, which are calculated by different risk measures, have better performance than that of the Taiwan Stock Exchange Capitalization Weighted Stock Index. (2) The estimates of the Value-at-Risk and Conditional Value-at-Risk predicted by different risk measures have crucial influence on the performance of the optimal asset allocation. When the confidence level is 95%, the bootstrap simulation model seems to have the best performance in the risk measures. In case of the 99% confidence level, equally weighted moving average model is the best one among the risk measures.

參考文獻


中文文獻
李吉元,「風險值限制下最適資產配置」,國立成功大學財務金融研究所碩士論文,民國九十二年六月。
阮建豐,「利用混合模型估計風險值的探討」,國立政治大學統計研究所碩士論文,民國九十一年六月。
洪幸資,「控制風險值下的最適投資組合」,國立政治大學金融研究所碩士論文,民國九十三年六月。
洪端禧,「動態隨機投資組合規劃 - 以亞太新興市場為研究對象」,國立高雄第一科技大學風險管理與保險所碩士論文,民國九十五年六月。

被引用紀錄


何佳豪(2011)。最適動態資產配置模型在投資組合之應用-以台灣五十成分股為例〔碩士論文,淡江大學〕。華藝線上圖書館。https://doi.org/10.6846/TKU.2011.01108
葉惠菁(2011)。MV 及 MCVaR 投資組合模型之績效評估-大中華區股市之實證研究〔碩士論文,淡江大學〕。華藝線上圖書館。https://doi.org/10.6846/TKU.2011.00895

延伸閱讀