本文關鍵詞：基于Copula函數-Asymmetric Laplace分布的金融市場風險度量與套期保值研究　出處：《華中科技大學》2013年博士論文　論文類型：學位論文

  更多相關文章： 金融市場風險 金融市場風險度量 套期保值 Copula函數 VaR

【摘要】：隨著金融全球一體化的發(fā)展，金融市場的復雜程度日益提高，防范金融風險已成為全社會的共識。加強金融系統(tǒng)風險防范和管理能力，提高市場轉移及消化吸收風險的能力，將是我國金融市場健康成長和發(fā)展的重要保障。金融秩序和金融運行環(huán)境的不斷改變，金融風險的產生、傳播、控制與管理等都日趨復雜，對金融市場風險的度量與管理的研究也更加重要和復雜。金融市場風險是最常見也是我國金融機構面臨的主要風險，但是對其的研究，一些傳統(tǒng)的基于正態(tài)、線性或波動性對稱等模型的研究已不再適用，很難充分地捕獲市場風險信息。這就需要不斷探索研究，給出更多適應現階段風險管理要求的理論模型研究及實證研究。 本文在分析現代金融風險管理理論的基礎上，總結了市場風險度量及期貨套期保值等方面的研究，指出了現有研究的不足，針對金融市場風險的復雜性，建立了基于非正態(tài)分布方法及非線性相關性模型的風險度量模型和套期保值策略模型，對金融市場風險的度量與套期保值進行了研究。主要從以下四個方面展開了主體部分的研究： （1）本文建立了基于Asymmetric Laplace（AL）分布的市場風險VaR與CVaR的度量模型。構建了市場風險VaR和CVaR度量的AL參數法和AL-MC法，并進行了比較研究。選取上證指數、日經225指數及SP500指數為研究對象，結合各股市的風險特征，給出了VaR和CVaR度量及其返回檢驗和準確性評價。結果表明，基于AL分布的風險度量模型能更好刻畫市場風險特征，能很好地度量市場風險。 （2）本文建立了動態(tài)風險VaR和CVaR度量的ARMA-GJR-AL模型。從相關性、波動性及殘差分布特征三方面考慮，研究了基于ARMA-GJR-AL模型的動態(tài)風險VaR和CVaR的度量。通過實證研究，給出了上海股市與紐約股市的市場風險預測及準確性檢驗，，研究了模型的有效性。結果表明，基于AL分布的動態(tài)風險度量模型更具合理性和適用性，能有效地度量風險。 （3）本文運用Copula函數技術來描述資產間的相關性結構，建立了金融資產組合的市場風險VaR和CVaR的度量和分配的Copula-AL模型，并對常用的基于多元統(tǒng)計分布的度量方法及基于OLS模型的風險分配方法進行了比較研究。選取上證指數和深圳成指的組合為例，計算了組合風險及其分配。結果表明，基于t-Copula-AL模型的VaR、CVaR法計算簡單準確，且能方便地進行風險分配。 （4）本文采用參數和非參數分布法來刻畫邊際分布特征，結合Copula函數技術來描述期現市場間的相關性，以CVaR最小化為目標函數，建立了基于靜態(tài)和動態(tài)Copula-CVaR的最優(yōu)套保比率度量模型，并對各模型進行了比較研究。以滬深300指數現貨和期貨為研究對象，建立了靜態(tài)和動態(tài)Copula-CVaR模型及OLS模型，在給定套保期限內，分析了各模型的套保費用，并給出了修正成本套保效率的比較分析。實證結果表明，考慮套保費用時，應選擇簡單易行的靜態(tài)套保策略，即使市場條件相同，也應據自身的費用情況選擇最優(yōu)套保策略。 本文的研究促進了金融市場風險度量、期貨套期保值、AL分布及Copula函數理論等方面的研究，具有很好的理論意義，同時對投資決策、經濟資本管理及風險管理等實踐活動也起到很好的幫助和借鑒作用。
[Abstract]:With the development of global financial integration, the complexity of the financial market is increasing, the prevention of financial risks has become the consensus of the whole society. To strengthen the financial system risk prevention and management ability, improve the ability of digestion and absorption and transfer market risk, will be our country financial market an important guarantee for the healthy growth and development of the constantly changing financial order and. The financial environment, financial risk, communication, control and management are becoming more and more complex, the research on financial market risk measurement and management is more important and complex. Financial market risk is the most common major risks facing China's financial institutions, but the research, based on the traditional normal the study of linear or volatility symmetry model is no longer applicable, it is difficult to fully capture the market risk information. This requires continuous exploration and research, give more suitable at this stage of the wind The theoretical model research and Empirical Study of risk management requirements.
Based on the analysis of modern financial risk management theory, summarizes the research of market risk measurement and Futures Hedging etc., points out the shortcomings of existing studies, aiming at the complexity of financial market risk, establish the risk of non normal distribution method and nonlinear correlation model measurement model and hedging strategy based on the model of the measure of financial market risk and hedging are studied. Mainly from the following four aspects of the research of the main part:
(1) is established in this paper based on Asymmetric Laplace (AL) model to measure market risk VaR and CVaR distribution. The construction parameters of the AL method and AL-MC method of VaR and CVaR to measure market risk, and a comparative study. Select the Shanghai index, Nikkei 225 index and SP500 index as the research object, combined with the characteristics of the risk the stock market, given the VaR and CVaR measure and return test and evaluation. The results show that the risk distribution of AL metric model can better describe the market risk based on the features, can be a good measure of market risk.
(2) this paper establishes the ARMA-GJR-AL model of dynamic risk VaR and CVaR metrics. From the correlation, volatility and residual distribution characteristics of three aspects, the research of dynamic risk measures of VaR and CVaR based on the ARMA-GJR-AL model. Through empirical research, market risk prediction accuracy and gives the Shanghai stock market and New York stock market test, research the validity of the model. The results show that the dynamic risk measurement model of AL distribution is more reasonable and based on the application, can effectively measure the risk.
(3) this paper uses the Copula function to describe the correlation between assets structure and technology, established the Copula-AL model to measure the market risk of CVaR and VaR combination of financial assets and distribution, and the measurement methods of multivariate statistical distribution and risk allocation method based on OLS model are studied based on the commonly used combination of Shanghai Composite Index and. Shenzhen stock market as an example, the calculation of portfolio risk and distribution. The results show that the t-Copula-AL model based on VaR, the CVaR method is simple and accurate, and can carry out risk distribution conveniently.
(4) this paper uses parametric and non parametric distribution method to describe the marginal distribution characteristics, combined with the Copula function to describe the correlation between the current market, with the goal of minimizing CVaR function, established the optimal hedging ratio of static and dynamic measurement model based on the Copula-CVaR rate, and the model is studied. The Shanghai and Shenzhen 300 and the stock index futures as the research object, established the static and dynamic Copula-CVaR model and OLS model in a given period, set limits, analysis of the cost of insurance set of each model, and gives the correct cost of hedging efficiency is analyzed. The empirical results show that considering the hedging costs, should choose the static hedging strategy is simple and, even if the market conditions are the same, should also choose the optimal hedging strategy according to their own expenses.
This paper promotes the financial market risk measurement, futures hedging, the research of AL distribution and Copula function theory, is of great theoretical significance, at the same time on investment decisions, economic capital management and risk management practices also play a very good help and reference.

【學位授予單位】：華中科技大學
【學位級別】：博士
【學位授予年份】：2013
【分類號】：F224;F831.51

【參考文獻】

相關期刊論文 前10條

1 胡援成,姜光明;上證綜指收益波動性及VaR度量研究[J];當代財經;2004年06期

2 曾健,陳俊芳;Copula函數在風險管理中的應用研究——以上證A股與B股的相關結構分析為例[J];當代財經;2005年02期

3 黃詒蓉;羅奕;;資本市場分形結構的理論與方法[J];當代財經;2006年03期

4 高全勝;金融風險計量理論前沿與應用[J];國際金融研究;2004年09期

5 田新時,劉漢中,李耀;滬深股市一般誤差分布(GED)下的VaR計算[J];管理工程學報;2003年01期

6 劉小茂,田立;VaR與CVaR的對比研究及實證分析[J];華中科技大學學報(自然科學版);2005年10期

7 陳守東,俞世典;基于GARCH模型的VaR方法對中國股市的分析[J];吉林大學社會科學學報;2002年04期

8 陳守東;胡錚洋;孔繁利;;Copula函數度量風險價值的Monte Carlo模擬[J];吉林大學社會科學學報;2006年02期

9 付勝華;檀向球;;股指期貨套期保值研究及其實證分析[J];金融研究;2009年04期

10 單國莉,陳東峰;一種確定最優(yōu)Copula的方法及應用[J];山東大學學報(理學版);2005年04期

相關博士學位論文 前1條

1 李夢玄;金融市場相依性Copula模型及實證研究[D];華中科技大學;2009年

本文編號：1420812