臺灣博碩士論文加值系統

本論文採取2001年4月10日到2006年4月30日共1,256筆日資料探討其樣本外的避險績效。研究標的包括：台灣加權股價指數現貨(TXs)、電子類指數現貨(TEs)、金融類指數現貨(TFs)、台指期(TX)、小台期(MTX)、電子期(TE)、金融期(TF)。本文利用兩種策略來估計平均避險比率與避險後的績效，第一種策略是使變異數最小化的避險策略，包含了naïve、 OLS、BI-GARCH、TGARCH、和ECM模型。第二種策略是考慮下檔風險最小的方式，並以LPM模型來衡量。估計期間分為100天與200天兩種，避險期間則有5天、10天、20天三種。實證結果顯示：
1.在第一種策略中，不論採用何種期貨指數，都是GARCH(1,1)的績效表現最佳，而天真模型表現最差。
2.在第一種策略中，當低階動差模型採用目標報酬率為所有現貨報酬的平均時，績效表現優於目標報酬率為零時。
3.平均而言，策略一的避險績效略高於策略二的績效表現。
4.考慮估計期間與避險期間下，本文發現無論採用何種期貨指數，隨著期間的增長則績效表現越佳。
5.把四種期貨一起比較，本研究發現小台指的績效表現低於台指期，這是因為台指期的交易量大，且流動性較佳的緣故。

This study investigated the out-of-sample hedging effectiveness for 1,256 observations between 10 April 2001 to 30 April 2006 for Taiwan futures market. The underlying assets include Taiwan weighted stock index (TXs), electronic sector index (TEs), financial sector index (TFs), Taiwan stock index futures (TX), mini Taiwan stock index futures (MTX), electronic sector index futures (TE), and financial sector index futures (TF). Two strategies are adopted to estimate the average of hedge ratios. The associated hedging effectiveness are also calculated. The first strategy focuses on examining minimum variance by applying the naïve, OLS, BI-GARCH, TGARCH, and ECM. The second strategy aims to minimize the downside risk by adopting LPM model. All data were collected and transferred to returns with the time expansions of 100-days and 200-days. The hedging periods are 5-days, 10-days, and 20-days.

By applying the first strategy, the hedging effectiveness of GARCH (1,1) performs best while naïve performs worst. As to the second strategy, the performance from LPM(c=μ) is larger than that from LPM(c=0). In average, the hedging effectiveness of the first strategy is usually larger than that of the second strategy.

When considering the time expansion, no matter which indices were adopted, hedging strategies perform better with increasingly estimated period and hedging period. Overall, it seems that the complicated models, such as GARCH(1,1) and ECM, would result in better hedging effectiveness. It is worth noting that the hedging effectiveness in MTX is lower than that in TX for all hedging models. This may be explained by the fact that the contract value of MTX is lower and the liquidity is better than that of TX.

Abstract v
Acknowledgement vii
Table of Contents viii
List of Tables x
List of Figures xi
Chapter 1 Introduction 1
1.1 Background and Motivation 1
1.2 Purposes of Study 2
1.3 Composition of Study 2
Chapter 2 Literature Review 4
2.1 Stock Index Futures 4
2.2 Review of Hedging Theory 8
2.2.1 Traditional Hedging Theory 9
2.2.2 Selective Hedging Theory 10
2.2.3 Portfolio Hedging Theory 11
2.3 Lower Partial Moment Theory 13
2.4 Relevant Literature Research 15
2.4.1 Foreign Empirical Results 15
2.4.2 Domestic Empirical Results 18
Chapter 3 Research Methodology 22
3.1 Data Source 22
3.2 Unit Root Test and Cointegration Test 24
3.2.1 Augmented Dickey-Fuller Test 24
3.2.2 Cointegration Test 25
3.3 Residual Test 27
3.3.1 Autocorrelation Residual Test 27
3.3.2 ARCH Effect Test 28
3.4 Hedging Strategy and Model 29
3.4.1 Hedging Strategy 29
3.4.2 Naïve Model 31
3.4.3 OLS Model 32
3.4.4 GARCH Model 32
3.4.5 ECM Model 36
3.5 Hedging Effectiveness 37
Chapter 4 Empirical Analysis 38
4.1 Descriptive Statistic 38
4.2 Unit Root and Cointegration Test 41
4.3 Arch Effect 44
4.4 Out-of-Sample Analysis 45
4.4.1 Hedge Ratio 45
4.4.2 Hedging Effectiveness 48
Chapter 5 Conclusive Remarks 51
5.1 Conclusions 51
5.2 Suggestions 52
References 53
Curriculum Vitae 55

1. Anderson, R. W., B. Danthine, and J. Pierre, "Cross hedge," Journal of Political Economy, 89(6), pp. 1182-1197, 1981.
2. Ghosh, A., “Hedging with stock index futures: estimation and forecasting with error correction model, introduction," Journal of Futures Markets, 9(4), pp. 457-475, 1993.
3. Chakraborty, A., and J. Barkoulas, "Dynamic futures hedging in currency markets," European Journal of Finance, 5(4), pp. 299-314, 1999.
4. Cecchetti, S. G., R. Cumby, and S. Figlewski, "Estimation of the optimal futures hedge," Review of Economics and Statistics, 70(4), pp. 623-30, 1988.
5. Chiu, C. L., P. Wu, C. Chen, and W. Cheng, "Hedging with floor-traded and E-mini stock index futures," Quarterly Journal of Business and Economics, 44, pp. 49-68, 2005.
6. Byström, H., "The hedging performance of electricity futures on the Nordic power exchange," Applied Economics, 35(1), pp. 1-11, 2003.
7. Phil, H., “Stock index futures hedging: hedge ratio estimation, duration effects, expiration effects and hedge ratio stability,” Journal of Business Finance and Accounting, 23(1), pp. 63-78, 1996.
8. Howard, C., and L. D'Antonio, "A risk-return measure of hedging effectiveness,'' Journal of Financial and Quantitative Analysis, 19, pp. 101-109, 1984.
9. Junkus, J. C., and C. Lee, ”Use of three stock index futures in hedging decisions,” Journal of Futures Markets, 5(2), pp. 231-237, 1985.
10. Donald, L., and T. Kuen, “Fractional cointegration and futures hedging,” The Journal of Futures Markets, 9(4), pp. 457-475, 1998.
11. Park, T. H., and L. N. Switzer, “Bivariate GARCH estimation of the optimal hedge ratios for stock index futures: a note,” The Journal of Futures Markets, l5, pp. 61-67, 1995.
12. McGlenchy, P., and P. Kofman, "Structurally sound dynamic index futures hedging," Econometric Society, 21(3), pp. 12-24, 2004.
13. Yeh, S. C., and G. L. Gannon, “Comparing trading performance of the constant and dynamic hedge models: a note,” Review of Quantitative Finance and Accounting, 14, pp. 155-160, 2000.