- 學位論文
以波動擇時架構分析 不同再平衡策略之績效差異
The Performance between Different Rebalancing Strategies under Volatility Timing Framework
摘要
本研究旨在探討波動擇時策略(volatility timing strategy)架構下，如何選擇最適之再平衡策略與估計期間資料筆數，因此本研究參考Liu (2009)作為基礎，考量資產配置下，利用E-mini S P 500期貨、美國10年期公債期貨與黃金期貨所形成之投資組合，在極小化風險的條件下以靜態模型、Bi-RGARCH模型與DCC-GARCH模型估計最適投資組合權重，並運用Sharpe Ratio、Sortino ratio、Omega ratio與經濟價值(Economic Value)，比較不同再平衡頻率與估計期間資料筆數對投資組合績效之影響，最後在考量交易成本下，探討是否有較佳的再平衡策略。 本研究結果發現在不考慮交易成本且每日再平衡的條件下，無論是否有放空限制，使用 Bi-RGARCH模型所建構之投資組合的績效均為最高，而每週再平衡的條件下則以DCC-GARCH模型有較佳的表現，其結果顯示每日再平衡的條件下包含已實現波動度的Bi-RGARCH模型可以凸顯出運用高頻資料於投資組合管理的效益，而較短的估計期間資料筆數其隱含的交易成本較低且有更佳的效益，此外本研究根據實務與文獻提出三種不同的再平衡策略，實證結果指出在考慮交易成本的情況下，策略二，即當權重變化超過一定門檻值時才進行再平衡之績效最為優異，為較佳之再平衡策略。
並列摘要
This study aims to investigate how to choose the most appropriate rebalancing strategy and estimation window under the framework of volatility timing strategy. Therefore, this study refers to Liu (2009) as the basis. Considering asset allocation, we use E-mini S P 500 futures, Treasury bond futures and gold futures to construct portfolios. Under the condition of minimizing volatility , static model, Bi-RGARCH model and DCC-GARCH model are used to estimate the optimal portfolio weights, and Sharpe Ratio, Sortino ratio, Omega ratio and Economic Value are used to compare the impact of different rebalancing frequencies and estimation windows on portfolio performance. The results of this study show that without considering transaction costs and daily rebalancing, the portfolios constructed by the Bi-RGARCH model have a better performance whether there is a short restriction. The DCC-GARCH model performs better while the portfolios are rebalanced weekly. These results show that under the condition of daily rebalancing, the Bi-RGARCH model, which involves realized volatility can highlight the benefits of high-frequency data in portfolio management, and the shorter estimation window has lower transaction costs and better performance. Additionally, this study proposes three different rebalancing strategies. The empirical results show that taking into account different transaction costs, the second strategy, that is, rebalance when the weight change exceeds a certain percentage has the best performance, which is considered as the better rebalancing strategy.