美國存託憑證與股票指數之不對稱外溢效果：次貸風暴的實證

The Spillover and Leverage Effects between ADRs and Underlying Stocks Indexes – the Empirical Study of Sumprime Mortgage Crisis.

蔡海倫(Hai-Lun Tsai)

指導教授 ： 秦宗春 ； 陳若暉

中原大學/商學院/國際貿易研究所/碩士(2008年)

https://doi.org/10.6840/cycu200900466

摘要

摘要
本研究探討次級房貸期間15個國家的27支美國存託憑證與各股市指數的影響，觀察期間為2000年1月至2008年2月。利用GARCH-ARMA 及EGARCH-ARMA分別討論報酬與波動性之外溢效果及槓桿效果是否存在於ADR市場及國際股票市場。將樣本分成產業別與地區別來討論美國存託憑證與相對國內股價指數的報酬與波動外溢效果。就產業而言，金融產業的國內股價指數明顯地受美國存託憑證股價報酬影響。值得注意地是，金融產業指數與美國存託憑證具有顯著雙向波動影響。就地區性而言，實證結果顯示歐洲、亞洲、及太平洋洲國家的股價指數與美國存託憑證有正向關係存在。歐洲和亞洲市場股價指數與美國存憑證之間有相對高地雙向波動外溢效果存在。
本文另採用調適性網路模糊推論系統ANFIS，其結合模糊理論的推論及類神經理論的學習功能，來預測美國存託憑證的未來價格。與GARCH-ARMA模型相比，調適性網路模糊推論系統的預測結果表現出極佳的預測能力。

關鍵字

外溢效果 ； EGARCH-ARMA模型 ； GARCH-ARMA模型 ； 槓桿效果 ； 調適性網路模糊推論系統 (ANFIS) ； 美國存託憑證

並列摘要

Abstract
The capital market’s reaction to the subprime mortgage crisis is analyzed in this paper by analyzing 27 American Depository Receipts (ADRs) from 15 countries and their underlying indexes in the period of the beginning of 2000 to Feb. 2008. The Generalized Conditional Heteroscedasticity-Autoregressive Moving Average (GARCH-ARMA) and the Expotentially Generalized Conditional Heteroscedasticity-Autoregressive Moving Average (EGARCH-ARMA) are employed to examine the existence of return and volatility spillover and leverage effects among ADR markets and foreign stock markets.
The sample is classified by industry and area to examine the transmission between ADR markets and foreign stock markets. From industry perspective, the significant return spillover effects from ADR markets to foreign underlying stock markets for financial industries are identified. The empirical results also suggest the financial industries have relatively significant bi-lateral relationships of volatility spillover effects. The evidence reveals that European, Asian, and Oceania markets have positive returns transmitted from lagged ADRs returns to the local stock indexes returns. Notably, the empirical results also indicate that significant bi-directional spillover and leverage effects of return and volatility exist in the European and Asian markets.
Moreover, this paper uses Adaptive Neuro-Fuzzy Inference System (ANFIS) equipped with fuzzy inference and self-learning capability in neuro network to predict the future price of ADR. The results show that ANFIS has a better forecast than GARCH-ARMA model.

並列關鍵字

GARCH-ARMA model ； EGARCH-ARMA model ； Spillover effect ； Leverage effect ； ADR ； ANFIS

參考文獻

Alaganar, V. T. and Bhar, R. (2001), “Diversification Gains from American Depositary Receipts and Foreign Equities: Evidence from Australian Stocks”, Journal of International Financial Markets, Institutions and Money, Vol.11, p97-113.

Baba, N. and M. Kozaki (1992), “An Intelligent Forecasting System of Stock Price Using Neural Networks”, IEEE International Joint Conference on Neural Networks, Vol.1, p371-377.

Bollerslev, T. (1986), “Generalized Autoregressive Conditional Heteroscedasticity”, Journal of Econometrics, Vol. 31, p307-327.

Brooks, Chris and Henry, Olan T. (2000), “Linear and Non-Linear Transmission of Equity Return Volatility: Evidence from The U.S., Japan and Australia”, Economic Modelling, Vol.17, p497-513.

Chang, Zhipeng, Liu, Liping and Li, Zhiping (2007), “Prediction of Amount of Imports Based on Adaptive Neuro-Fuzzy Inference System”, IEEE International Conference on Intelligent Pervasive Computing, p437-440.