臺股指數期貨保證金估計模型及結構比之研究

On the Methodologies of Margin-Setting and the Requirements for Initial, Maintenance and Clearing Margins – The Case of TAIFEX Stock Index Futures

施宗佐(Tsung-Tzuo Shih)

指導教授 ： 李存修

國立臺灣大學/管理學院/財務金融學研究所/碩士(2009年)

https://doi.org/10.6342/NTU.2009.10737

摘要

我國期貨市場漸趨成熟，如何降低市場違約風險、達到履約保證要求的同時，吸引更多潛在交易人進行交易與避險，一直是重要議題，因此，研究重點在於台股期貨之保證金水準與保證金結構比之訂定。

研究以簡單移動平均模型、指數加權移動平均、一般化自我迴歸條件異質變異數模型以及極值理論模型對臺股期貨做回溯測試，評估四種模型在保證金水準、穿透率、資金效率、保證金調整之優劣，發現簡單移動平均模型為最適模型。

而在保證金結構比方面，研究將現行保證金結構比與國際主要交易所常用指數保證金結構比，以此兩種結構比進行回溯測試，發現在違約風險無虞情況下，變更臺股期貨之保證金結構比，將增加資金效率。

關鍵字

保證金結構比 ； 簡單移動平均 ； 指數加權移動平均 ； 一般化自我迴歸條件異質變異數 ； 極值理論

並列摘要

As the Taiwanese future market grows mature, how to reduce the market default risk and make sure traders secure their deposits, and at the same time, attract more potential traders has been an important issue. As a result, this study focuses on not only the margin-setting methodologies, but also the requirements for clearing, maintenance, initial margins of TAIEX futures.

In this study, four margin-setting methodologies, namely, one using the simple moving average, one using exponential weighted moving average, one using GARCH approach, and the other using extreme value theory, are conducted and evaluated in terms of the margin level, failure rate, fund efficiency and the margin adjustment through back testing. The results indicate that the simple moving average outperform other models.

As for determining the requirements for the clearing, maintenance, and initial margins, we compare the current requirements used in Taiwan Future Exchange and the prevailing requirements used in the international exchanges through back testing. It shows that if no default risk is assumed, changing the ratio for TAIFEX future will increase the fund efficiency.

並列關鍵字

Requirements for clearing, maintenance, and initial margins ； Simple moving average ； exponential weighted moving average ； GARCH(1,1) ； Extreme value theory

參考文獻

4. 張雅惠(2007), 肥尾模型的波動性預測, 淡江大學財務金融學系碩士在職專班碩士論文.

6. Bali, Turan G., 2003, “An Extreme Value Approach to Estimating Volatility and Value at Risk,” Journal of Business, 76(1).

7. Barone-Adesi, G. and Kostas, G., 2000, “Non-parametric VaR techniques. Myths and Realities,” Working paper.

8. Bollerslev, T. 1986, "Generalized Autoregressive Conditional Heteroskedasticity," Journal of Econometrics, 31, 307-327.

9. Booth, G.G., Broussard, J.P., Martikainen, T., Puttonen, 1997, "Prudent Margin Levels in the Finnish Stock Index Futures Market," Management Science, 43, 1177-1188.