臺灣博碩士論文加值系統

本文考量如何有效估計由具共整合(co-integration)效應資產所組成之統計套利(statistical arbitrage)投資策略的期望報酬，在該投資策略之報酬率滿足獨立且同分布的假設下，提出一具有封閉遞迴表達式的估計方法。應用本文所提出的方法可大量縮短傳統上利用模擬方法估計投資策略期望報酬的計算時間。在假設投資策略之報酬率滿足常態、混和常態(normal mixture)或高斯核分佈(Gaussian kernel distribution)等三種情況下，模擬結果顯示該方法皆可快速地得到精確的估計結果。此外，在加入交易成本考量時，應用本文所提出的估計方法亦有助於設立最佳共整合統計套利投資策略。實證研究方面，本文以台灣中型一百指數在2010年1月至2014年3月17日期間的資料為對象，觀察所提出之最佳投資策略的獲益能力。

This study proposes an efficient estimate for the expected returns of co-integration based statistical arbitrage trading strategies. If the returns of the trading strategy are independent and identically distributed, a recursive closed-form representation of its expectation is established. The proposed estimation is capable of significantly saving much more computational effort than traditional simulation based methods. Numerical results indicate that the proposed method obtains accurate estimates effectively if the returns of a co-integration based statistical arbitrage trading strategy are normal, normal mixture or Gaussian kernel distributed. In addition, under consideration of transaction costs, the optimal co-integration based statistical arbitrage trading strategy is obtained effectively by the proposed estimate. The empirical study employs the data of Taiwan Mid-Cap 100 Indices during the period from January 1, 2010 to March 17, 2014 to investigate the performance of the proposed optimal trading strategy with transaction costs.

中文摘要 ii
英文摘要 iii
1 緒論 1
1.1 研究背景與動機 1
1.2 研究方法與目的 2
2 文獻回顧 4
2.1 配對交易 4
2.2 共整合套利 4
3 主要結果 6
4 模擬研究 11
4.1 常態模型 11
4.2 混和常態模型 12
4.3 高斯核分布模型 14
5 實證分析 15
6 結論 16
附錄 18
參考文獻 23

[1] Alexander, C. (1999). Optimal hedging using cointegration. Philo-
sophical Transactions of the Royal Society A, 357, 2039-2058.
[2] Alexander, C. and Dimitriu, A. (2002). The cointegration alpha:
enhanced index tracking and long-short equity market neutral
strategies. ISMA Discussion Papers in Finance.
[3] Chiu, M.C. and Wong, H.Y. (2011). Mean-variance portfolio se-
lection of cointegrated assets. Journal of Economic Dynamics &
Control, 35, 1369-1385.
[4] Duan, J.C. and Pliska, S.R. (2004). Option valuation with co-
integrated asset prices. Journal of Economic Dynamics & Control,
28, 727-754.
[5] Granger, C.W.J. (1981). Some properties of time series data and
their use in econometric model speci cation. Journal of Economet-
rics, 16, 121-130.
[6] Tsay, R.S. (2010). Analysis of nancial time series(3th ed.). New Jersey: John Wiley & Sons.