以LTVECM模型探討美國存託憑證與其標的股價格動態傳輸之門檻與預測效果

Using LTVECM to Analyze the Forecasting of American Depositary Receipts and Their Underlying Securities

10.30114/CGJHSS.200810.0004

王淳玄(Chun-Hsuan Wang)洪裕發(Yu-Fa Hong)

水準門檻誤差修正模型 ； 共整合 ； 預測成效 ； Level threshold error correction model ； Cointegration ； Forecasting performance

長庚人文社會學報

1卷2期（2008 / 10 / 01）

283 - 310

繁體中文

本研究利用誤差修正模型(Vector Error Correction Model; VECM)、門檻誤差修正模型(threshold VECM; TVECM)與水準門檻誤差修正模型(level TVECM; LTVECM)以分析美國存託憑證與其標的股股價間動態傳輸之門檻效果，並利用拔靴複製模擬法比較模型之預測效果。研究樣本爲目前台灣到美國NYSE﹑AMEX或NASDAQ發行ADR之公司，包括台積電(TSM)、日月光(ASX)、旺宏(MXICY)、聯電(UMC)、友達(AUO)與矽品(SPIL)。初步研究結果發現，所有變數均爲I(1)且各成對樣本均存在共整合關係。 再者，由於許多財務與經濟變數均具有非線性特性。本研究採用Hansen and Seo(2002)之SupLM檢定以檢驗模型門檻效果。結果顯示，部份樣本存在模型門檻效果。另一方面則利用RMSFE評估準則以比較模型之相對預測效果。研究結果發現，當門檻檢定結果顯示模型不存在門檻效果時，VECM之預測效果優於TVECM與LTVECM之預測效果。相反地，當門檻檢定結果顯示模型存在門檻效果時，LTVECM之預測效果則優於VECM與TVECM之預測效果。此結果說明門檻檢定結果與預測效果評估具一致性。另外，於非線性模型中，不同區間存在不同長期均衡關係之假設更能符合實際資料之特性，亦即以LTVECM較能提高其預測效果。

In this study, we employ vector error correction model (VECM), threshold vector error correction model (TVECM) and level threshold error correction model (LTVECM) to analyze the threshold effect of dynamic transmission between the American Depositary Receipts (ADRs) and their underlying stocks. In addition, the bootstrapping simulations are utilized for evaluating the forecasting performance of models. All ADRs included in this study are ASX, AUO, MXICY, SPIL, TSM and UMC. The initial outcomes show that all variables are integrated of degree one and those ADRs are integrated with their underling shares. Furthermore, many financial and economic variables are non-linear. We apply SupLM test to investigate the possibility that the models studied here have threshold effect. The results suggest that some data are suitable for threshold models. On the other hand, we use the RMSFE for the comparison of forecasting performance of models. Based on simulation experiments, the forecasting performance of the VECM outperforms the TVECM and LTVECM as the model does not have threshold effect, while the forecasting performance of the LTVECM outperforms the VECM and TVECM as the model has threshold effect. These outcomes demonstrate the consistency of the threshold effect tests and the forecasting performance evaluation. Therefore, LTVECM may be more suitable for our data with non-linearity.

1. Balke, N. S.,Fomby, T. B.(1997).Threshold Cointegration.International Economic Review,38,627- 645.
2. Chen, S. Y.,Chou, L.,Yang, C.(2002).Price Transmission Effect between GDRs and their Underlying Stocks-Evidence from Taiwan.Review of Quantitative Finance and Accounting,19(2),181-213.
3. De Gooijer, J. G.,Vidiella-i-Anguera, A.(2005).Estimating Threshold Cointegrated Systems.International Journal of Forecasting,20,237-253.
4. De Gooijer, J. G.,Vidiella-i-Anguera, A.(2004).Forecasting Threshold Cointegrated Systems.Economics Bulletin,3(8),1-7.
5. Dickey, A. D.,Fuller, A.(1981).Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root.Econometrica,59,1057-1072.
6. Ely, D.,Salehizadeh, M.(2001).American Depositary Receipts an Analysis of International Stock Price Movements.International Review of Financial Analysis,10,343-363.
7. Engle, R. F.,Granger, C. W. J.(1987).Cointegration and Error Correction: Representation, Estimation, and Testing.Econometrica,55,251-276.
8. Hansen, B. E.,Seo, B.(2002).Testing for Two-Regime Threshold Cointegration in Vector Error Correction Models.Journal of Econometrics,110,293-318.
9. Jiang, C. X.(1998).Diversification with American Depository Receipts: the Dynamics and the Pricing Factors.Journal of Business Finance and Accounting,25,683-699.
10. Johansen, S.(1988).Statistical Analysis of Cointegration Vectors.Journal of Economic Dynamics and Control,12,231-254.
11. Johansen, S.,Juselius, K.(1990).Maximum Likelihood Estimation and Inference on Cointegration with Application to the Demand of Money.Oxford Bulletion of Economics and Statistics,52(2),169-210.
12. Kato, K.,Linn, S.,Schallheim, J.(1991).Are there Arbitrage Opportunities in the Market for American Depositary Receipts.Journal of International Financial Markets,1,73-89.
13. Kim, M.,Szakmary, A. C.,Mathur, I.(2000).Price Transmission Dynamics between ADRs and their Underlying Foreign Securities.Journal of Banking and Finance,24,1359-1382.
14. Kwiatkowski, D.,Phillips, P. C. B.,Schmidt, P.,Shin, Y.(1992).Testing the Null Hypothesis of Stationarity Against the Alternative of a Unit Root.Journal of Econometrics,54,159-178.
15. Miller, D. P.,Morey, M. R.(1996).The Intraday Pricing Behavior of International Dually Listed Securities.Journal of International Financial Markets, Institutions and Money,6,79-89.
16. Park, J.(1990).University of IOWA.
17. Park, J.,Tavakkol, A.(1994).Are ADRs a Dollar Translation of their Underlying Securities? The Case of Japanese ADRs.Journal of International Financial Markets,4,77-87.
18. Phillips, P.,Perron, P.(1988).Testing for Unit Root in Time Series Regression.Biometrika,75,335-346.
19. Siklos, P. L.,Granger, C. W. J.(1997).Regime-Sensitive Cointegration with an Application to Interest-Rate Parity.Macroeconomic Dynamics,1,640-657.
20. Solnik, B.(1996).International Investments.New York:Addision-Wesley.
21. Tong, H.,Chen, C.H.(ed.)(1978).On a Threshold Model in Pattern Recognition and Signal Processing.Amsterdam:Sijhoff and Noordhoff.
22. Tsay, R. S.(1998).Testing and Modeling Multivariate Threshold Models.Journal of the American Statistical Association,93,1188-1202.
23. Wahab, M.,Khandwala, A.(1993).Why not Diversify Internationally with ADRs?.Journal of Portfolio Management,19,75-82.
24. Wang, C.,Lin, C. A.(2005).Using Threshold Cointegration to Examine Asymmetric Price Adjustments between ADRs and their Underlying Securities-the Case of Taiwan.South African Journal of Economics,73(3),449-462.