謝德宗Der-Tzon, Hsieh臺灣大學:經濟學研究所應文逡Ing, Wen-ChiunWen-ChiunIng2010-05-052018-06-282010-05-052018-06-282009U0001-2008200914594500http://ntur.lib.ntu.edu.tw//handle/246246/179530本研究利用台灣卓越 50 指數股票型基金(Taiwan 50 ETF)代替指數現貨作為指數期貨和指數選擇權的套利工具，解決過去使用一籃子股票複製大盤所造成的問題，諸如：高額交易成本、流動性較低等。研究實證分析二００九年一月至五月，台指選擇權與台指期貨市場間之套利關係，考量交易成本等因素後，我們發現可套利機會幾乎不存在。００三年六月至二００九年六月，ETF 50對指數現貨和ETF 50對指數期貨皆無共整合關係存在。但若取資料二００七年一月至二００九年六月，則存在共整合關係，且ETF 50、指數現貨和指數期貨三者間，除指數期貨對指數現貨外，皆存在Granger因果關係。後經由線性迴歸得到1:32的指數期貨對ETF 50 價格變動關係，再取資料二００九年四月至二００九年六月作套利實證得到相當16%年報酬率。若使用財務槓桿，則可得到大約24%年報酬率。With transaction cost, there’s almost no arbitrage opportunity exists between TAIEX options and futures for non-market makers during the data period of Jan., 2009 to May, 2009.or ETF 50, futures and TAIEX index, ETF 50-Futures and ETF 50-TAIEX index do not have cointegration relationship for the period of Jun. 30, 2003 to Jun. 30, 2009. However, futures-TAIEX index, ETF 50-Futures and ETF 50-TAIEX index all exist cointegration relationship at 1% significant level for the period of Jan. 1, 2007 to Jun. 30, 2009. ETF 50, futures and TAIEX index all have Granger causality relationship except Future-TAIEX index, Jan. 2007 to Jun. 2009.inear regression was employed for the best substitute ratio. One unit price of TAIEX futures changes will cause 32 units price change of ETF 50. For arbitrage trade, one contract of futures has to be hedged with 32 contracts of ETF 50.erforming arbitrage trading from Apr. to Jun., 2009, an annual profit rate of around 16% was obtained. If leverage is deployed, then the annual profit rate may go up to ~24%.口試委員會審定書 iCKNOWLEDGEMENTS ii文摘要 iiiBSTRACT ivONTENTS vIST OF FIGURES viiIST OF TABLES viiihapter 1 Introduction 1.1 Background 1.2 Purpose of this Study 4.3 Thesis Organization 4hapter 2 Literature Review 6hapter 3 Methodology and Data Source 10.1 Data Source 10.2 Cost of Carry Model 12.3 The Put-Call-Futures Parity Condition 12.4 Time series data 14.5 Cointegration test 16.6 Causality and error correction model 17.6.1 Granger causality test 17.6.2 Error correction model 18hapter 4 Empirical Results 19.1 Data Descriptions 19.2 Futures vs. Options 21.2.1 Transaction Cost 21.2.2 Data Analysis 21.3 Index vs. Futures 22.3.1 Correlations 22.3.2 ADF - Unit Root Tests 23.3.3 Cointegration 25.3.4 Granger causality test 28.3.5 ETF 50 substitute ratio for arbitrage 28.3.6 Empirical Results for Futures-ETF Arbitrage 30hapter 5 Conclusions 33EFERENCE 35application/pdf969789 bytesapplication/pdfen-US指數期貨指數選擇權指數股票型基金套利index optionsindex futuresETFarbitragethesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/179530/1/ntu-98-P93323028-1.pdf