臺灣博碩士論文加值系統

傳統風險值的衡量忽略關聯市場之間的連動情形，本研究使用條件風險值(CoVaR)，其加入美股S&P作為條件值，試圖藉此捕捉新興市場與美股的連動性冀望估計出符合信賴區間下，更貼近實際情況之預期損失。本研究以「亞洲四小龍」及「靈貓六國(CIVETS)」等十個新興國家股市作為研究對象，研究期間為2000年07月28日至2022年12月30日止，為解決各市場間存在的時差問題，除時區接近的哥倫比亞外，其餘皆使用往前一期的美國股價指數和當期的新興市場指數進行研究。在研究中分別計算服從t分配之風險值、以單變量GARCH模型所計算之風險值，以及考量雙變量DCC GARCH模型估計出美股與新興市場間的動態相關係數，並據此所計算出之條件風險值；針對全研究區間、次貸危機期間以及疫情期間等三個時間段，使用三種信賴水準進行回測。實證結果顯示：新興市場與美股之間皆存在一定程度的動態相關，其中相關性最強國家為韓國KOSPI，其次為新加坡STI；而根據風險值估計之結果，雖於長期而言，條件風險值CoVaR提供較傳統風險值更為保守的風險估計，但在遇到極端事件或危機時，相較於傳統風險值的衡量，於大多數的新興市場中，條件風險值更能捕捉市場連動所存在的預期損失。

The conventional measurement of risk overlooks the interconnectedness between related markets. This study employs conditional value-at-risk (CoVaR), incorporating the US stock market's S&P as a conditioning variable, to capture the linkage between emerging markets and the US stock market, aiming to estimate expected losses that align more closely with real markets within reliable confidence intervals. The research focuses on the stock markets of ten emerging countries, including the 'Four Asian Tigers' and 'CIVETS.' The study period spans from July 28, 2000, to December 30, 2022. To address the issue of time differences, except for Colombia due to its close time zone, the others use lagged US stock index data and concurrent emerging market index data for analysis.

During the study, we calculate three risk values separately: the risk value following the t-distribution, the risk value derived from the univariate GARCH model, and the conditional risk value (CoVaR) based on the bivariate DCC GARCH model's estimated dynamic correlation coefficient between US stocks and emerging markets. Backtesting is conducted at three confidence levels for the entire research period, the subprime mortgage crisis, and the pandemic period.

The empirical results demonstrate the existence of a certain degree of dynamic correlation between emerging markets and the US stock market. Notably, South Korea's KOSPI exhibits the strongest correlation, followed by Singapore's STI. Regarding the risk value estimations, CoVaR generally provides a more conservative risk estimate compared to the traditional risk values, especially during extreme events or crises. In most emerging markets, CoVaR outperforms the traditional risk measurement in capturing the expected loss associated with market interdependencies.


Keywords: Value at Risk (VaR), Conditional Value at Risk (CoVaR), Univariate GARCH model, Bivariate DCC GARCH model, Emerging markets.

目錄
摘要 I
ABSTRACT II
致謝 III
目錄 IV
表目錄 V
圖目錄 VI
第一章、緒論 1
第一節、研究動機 1
第二節、研究目的 6
第三節、研究架構 6
第二章、文獻回顧與背景探討 7
第一節、新興市場股市之背景介紹 7
第二節、風險值與條件風險值相關文獻 13
第三章、研究方法 17
第一節、連續複利 (對數)收益率與單根檢定 17
第二節、風險值與條件風險值之定義 19
第三節、風險值與條件風險值之實證模型 20
第四節、回測方法 25
第四章、實證結果 29
第一節、資料來源與基礎統計特性 29
第二節、樣本內估計參數 30
第三節、回測方法 35
第四節、回測結果 54
第五章、結論 56
參考文獻 57
附錄 60

參考文獻
Tobias, A., & Brunnermeier, M. K. (2016). CoVaR. The American Economic Review, 106(7), 1705.

Bernardi, M., Gayraud, G., & Petrella, L. (2015). Bayesian tail risk interdependence using quantile regression.

Basel Committee on Banking Supervision, (1996).Supervisory Framework for the Use of 'Backtesting' in Conjunction with the Internal Models Approach to Market Risk Capital Requirements. https://www.bis.org/publ/bcbs22.htm.

Bernardi, M., Maruotti, A., & Petrella, L. (2013). Multivariate Markov-Switching models and tail risk interdependence. arXiv preprint arXiv:1312.6407.

Bhanot, K., Burns, N., Hunter, D., & Williams, M. (2014). News spillovers from the Greek debt crisis: Impact on the Eurozone financial sector. Journal of Banking & Finance, 38, 51-63.

Christoffersen, P. F. (1998). Evaluating interval forecasts. International economic review, 841-862.

Cao, Z. (2013). Multi-covar and shapley value: A systemic risk measure. Banq. France Work. Pap.

Engle, R. F. (1982). Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. Econometrica: Journal of the econometric society, 987-1007.

Engle III, R. F., & Sheppard, K. (2001). Theoretical and empirical properties of dynamic conditional correlation multivariate GARCH.

Girardi, G., & Ergün, A. T. (2013). Systemic risk measurement: Multivariate GARCH estimation of CoVaR. Journal of Banking & Finance, 37(8), 3169-3180.

Jin, X. (2018). Downside and upside risk spillovers from China to Asian stock markets: A CoVaR-copula approach. Finance Research Letters, 25, 202-212.

Philippe Jorion.(2000). Value at Risk: The New Benchmark for Managing Financial Risk (2nd). McGraw-Hill Book Co.

Kupiec, P. H. (1995). Techniques for verifying the accuracy of risk measurement models (Vol. 95, No. 24). Division of Research and Statistics, Division of Monetary Affairs, Federal Reserve Board.

Mokni, K., & Mansouri, F. (2017). Conditional dependence between international stock markets: A long memory GARCH-copula model approach. Journal of Multinational Financial Management, 42, 116-131.

Mainik, G., & Schaanning, E. (2012). On Dependence Consistency of CoVaR and Some Other Systemic Risk Measures. RiskLab, Department of Mathematics, ETH Zurich.

Nelson, D. B. (1991). Conditional heteroskedasticity in asset returns: A new approach. Econometrica: Journal of the econometric society, 347-370.

Oh, D. H., & Patton, A. J. (2013). Simulated method of moments estimation for copula-based multivariate models. Journal of the American Statistical Association, 108(502), 689-700.

Trabelsi, N., & Naifar, N. (2017). Are Islamic stock indexes exposed to systemic risk? Multivariate GARCH estimation of CoVaR. Research in International Business and Finance, 42, 727-744.

Peña, D., Tiao, G. C., & Tsay, R. S. (Eds.). (2001). A course in time series analysis (Vol. 409). New York: Wiley.

Reboredo, J. C., & Ugolini, A. (2015). Systemic risk in European sovereign debt markets: A CoVaR-copula approach. Journal of International Money and Finance, 51, 214-244.

黃小菁 (2005)，DCC 多變量 GARCH 模型之風險值計算-G7 及臺灣等八國股市投資組合之實證研究，淡江大學財務金融學系碩士在職專班。

柯博倫 (2010)，風險值之估計-GARCH 模型之應用，國立台灣大學農業經濟學研究所。