本文選題：跳-擴散模型　切入點：均值、方差跳躍點　出處：《北方工業(yè)大學》2014年碩士論文　論文類型：學位論文

【摘要】：跳躍現(xiàn)象在經濟、金融、工業(yè)流程控制等領域普遍存在,對跳躍的把握程度直接影響到是否能夠避免不必要的損失以及能否駕馭該領域里有益的轉變時機。在金融資產價格序列中,研究人員是通過對資產價格的建模來研究這種跳躍特性的。在對股指收益率建模中,跳-擴散模型對收益率序列的擬合效果非常好,能很好地反應收益率序列的一些特征。 本文提出了一個新的跳躍點檢測算法,并將之運用到中國股票市場的實證分析中。跳躍點可分為均值跳躍點和方差跳躍點,針對其不同特性,我們選取V統(tǒng)計量檢測均值跳躍點,并構造了一個新統(tǒng)計量用來檢測方差跳躍點。 本文主要工作如下：首先,利用方差跳躍點前后收益率序列偏差變化顯著這個特性,本文構造了一個新的方差跳躍檢驗統(tǒng)計量--類偏差比(DLR)統(tǒng)計量,并給出了基于非對稱Poisson跳-擴散模型閾值確定的模擬方法。其次,本文根據(jù)新的方差跳躍檢驗統(tǒng)計量DLR和均值跳躍檢驗統(tǒng)計量V提出了一種新的跳躍點檢測算法。最后,本文將新算法應用到中國股票市場的實證分析中。 我們選取了中國股票市場中五個主要的股指數(shù)據(jù)(上證指數(shù),上證50指數(shù),深證綜合指數(shù),深證成份股指數(shù),滬深300指數(shù)),以2007年1月4日到2011年12月30日五年的股指日收盤價作為樣本數(shù)據(jù),利用本文提出的新方法對中國股票市場進行了實證分析。實證分析表明中國股票市場對于政府政策的依賴性較大,且受國際股市波動的影響較大；另外,我們還需要不斷地加強政府政策的監(jiān)管力度,建立健全的法律規(guī)章制度體系,確保外部環(huán)境正常和諧地運行,推進中國股票市場的創(chuàng)新性改革,加強和完善股票市場組織管理機構,從而使得中國股票市場更加成熟、穩(wěn)定。
[Abstract]:The phenomenon of jumping is widespread in the fields of economy, finance, industrial process control, etc. The degree of mastery of the jump has a direct bearing on whether unnecessary losses can be avoided and whether the beneficial timing of the transition in this area can be harnessed. In the modeling of stock index yield, the jump-diffusion model fits the yield series very well and can well reflect some characteristics of the return series. In this paper, a new jumping point detection algorithm is proposed and applied to the empirical analysis of Chinese stock market. The jump point can be divided into mean jump point and variance jump point. We select V statistic to detect mean jump point and construct a new statistic to detect variance jump point. The main work of this paper is as follows: firstly, a new variance jump test statistic, class deviation ratio (DLR), is constructed by using the characteristic of significant variation of return sequence deviation before and after variance jump point. A simulation method based on the threshold of asymmetric Poisson hop-diffusion model is given. Secondly, according to the new variance jump test statistic DLR and the mean jump test statistic V, a new jumping point detection algorithm is proposed. This paper applies the new algorithm to the empirical analysis of Chinese stock market. We have selected five main stock index data in Chinese stock market (Shanghai Stock Exchange Index, Shanghai Stock Exchange 50 Index, Shenzhen Stock Exchange Composite Index, Shenzhen Stock Exchange component Index. The Shanghai-Shenzhen 300 index is based on the daily closing price of the stock index for five years from January 4th 2007 to December 30th 2011. The new method proposed in this paper is used to analyze the Chinese stock market. The empirical analysis shows that the Chinese stock market depends heavily on the government policy and is greatly affected by the volatility of the international stock market. We also need to constantly strengthen the supervision of government policies, establish a sound legal and regulatory system, ensure the normal and harmonious operation of the external environment, and promote the innovative reform of China's stock market. Strengthen and perfect the organization and management of stock market, thus make Chinese stock market more mature and stable.
【學位授予單位】：北方工業(yè)大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：F832.51;F224

【參考文獻】

相關期刊論文 前4條

1 黃香,葉維彰,欒貽會,謝衷潔;跳躍點統(tǒng)計檢測的小波方法及其在金融匯率中的應用[J];北京大學學報(自然科學版);1997年03期

2 王建穩(wěn),肖春來;非對稱Possion跳——擴散模型的參數(shù)估計[J];數(shù)理統(tǒng)計與管理;2005年04期

3 王建穩(wěn);Possion跳—擴散模型的參數(shù)估計[J];數(shù)學的實踐與認識;2005年07期

4 秦磊;;股市跳躍點對股票價量變化的影響分析[J];商業(yè)時代;2011年07期

，

本文編號：1622673