【摘要】：在我國(guó)股票市場(chǎng)，股價(jià)常常在一定時(shí)期內(nèi)發(fā)生劇烈波動(dòng)，給個(gè)人投資者帶來(lái)不可預(yù)料的損失。Markowitz投資組合理論奠定了現(xiàn)代投資組合理論的基礎(chǔ)，然而投資組合只能有效降低非系統(tǒng)風(fēng)險(xiǎn)，對(duì)系統(tǒng)風(fēng)險(xiǎn)的規(guī)避卻無(wú)能為力。本論文以中小企業(yè)板為研究對(duì)象，對(duì)股票價(jià)格波動(dòng)過(guò)程中低點(diǎn)、高點(diǎn)的出現(xiàn)及出現(xiàn)的規(guī)律進(jìn)行了系統(tǒng)分析和深入研究。首先在對(duì)Markowitz投資組合理論和股票價(jià)格波動(dòng)理論的研究基礎(chǔ)上，以不同時(shí)長(zhǎng)股票階段成本作為參考值，在前提條件約束下，設(shè)計(jì)了一系列股票價(jià)格波動(dòng)過(guò)程中階段周期低點(diǎn)、高點(diǎn)回歸預(yù)測(cè)模型。然后采用點(diǎn)估計(jì)、區(qū)間估計(jì)等對(duì)符合采樣條件的樣本進(jìn)行檢驗(yàn)和比較，驗(yàn)證不同模型對(duì)價(jià)格極值進(jìn)行預(yù)測(cè)的真實(shí)性并得到最優(yōu)模型。最后應(yīng)用該模型進(jìn)行股票投資的模擬仿真操作，，對(duì)股票選擇和投資組合管理進(jìn)行對(duì)策研究，并分析了這種操作對(duì)系統(tǒng)風(fēng)險(xiǎn)的規(guī)避能力。在投資組合基礎(chǔ)上應(yīng)用該模型進(jìn)行股票處理的對(duì)策研究，對(duì)個(gè)人投資者有很好的指導(dǎo)意義。本論文的主要研究?jī)?nèi)容概括如下： 1．考慮到Markowitz投資組合理論在我國(guó)的實(shí)用性和它在系統(tǒng)風(fēng)險(xiǎn)降低方面的限制性，提出了基于股票價(jià)格波動(dòng)問(wèn)題的階段周期低點(diǎn)、高點(diǎn)回歸模型。 2．以中小企業(yè)板為背景，設(shè)計(jì)六種基于不同解釋變量的階段周期回歸預(yù)測(cè)分析模型。通過(guò)t檢驗(yàn)、F檢驗(yàn)、擬合優(yōu)度分析和殘差分析等各種統(tǒng)計(jì)檢驗(yàn)初步確定有效模型，然后針對(duì)多重共線性問(wèn)題對(duì)階段周期多元回歸預(yù)測(cè)模型進(jìn)行修正。 3．應(yīng)用不同預(yù)測(cè)方式，使用歷史數(shù)據(jù)對(duì)不同模型有效性和精確性進(jìn)行驗(yàn)證和分析，并得出修正后的階段周期多元回歸預(yù)測(cè)模型擬合效果最優(yōu)的結(jié)論。 4．針對(duì)最優(yōu)的階段周期回歸預(yù)測(cè)模型，提出三種不同模式的投資操作來(lái)進(jìn)行投資組合的實(shí)證研究，包括基于階段周期回歸預(yù)測(cè)模型的投資、不遵循建議的無(wú)作為投資和理想化的指數(shù)化投資。實(shí)證研究結(jié)果顯示，利用階段周期低點(diǎn)、高點(diǎn)回歸預(yù)測(cè)模型進(jìn)行選股和投資操作的投資組合，收益率遠(yuǎn)高于其他模式，并且能有效地規(guī)避系統(tǒng)風(fēng)險(xiǎn)。在基于階段周期回歸預(yù)測(cè)模型的操作過(guò)程中，對(duì)股票的選購(gòu)和拋售提出對(duì)策建議。
[Abstract]:In the stock market of our country, the stock price often fluctuates violently in a certain period, which brings unexpected losses to individual investors. Markowitz portfolio theory has laid the foundation of modern portfolio theory. However, the portfolio can effectively reduce the non-system risk, but can not avoid the system risk. This paper takes the SME board as the research object, carries on the systematic analysis and the thorough research to the stock price fluctuation process low point, the high point appearance and the appearance rule. Firstly, based on the research of Markowitz portfolio theory and stock price fluctuation theory, taking the stage cost of stock with different time and length as reference value, a series of periodic low points in the process of stock price fluctuation are designed under the constraints of preconditions. High point regression prediction model. Then point estimation and interval estimation are used to test and compare the samples that meet the sampling conditions to verify the authenticity of price extremum prediction by different models and to obtain the optimal model. Finally, the model is used to simulate and simulate the stock investment, and the countermeasures of stock selection and portfolio management are studied, and the ability of this operation to avoid the system risk is analyzed. On the basis of investment portfolio, the research on the countermeasures of stock processing based on this model has a good guiding significance for individual investors. The main research contents of this thesis are summarized as follows: 1. Considering the practicability of Markowitz portfolio theory in our country and its limitation in reducing system risk, a phase cycle low based on stock price volatility is proposed. High point regression model. 2. Taking the SME board as the background, six models of stage cycle regression analysis based on different explanatory variables are designed. Through t test, F test, goodness of fit analysis, residual analysis and other statistical tests, the effective model is preliminarily determined, and then the stage periodic multivariate regression prediction model is modified for multi-multiple collinear problems. 3. The validity and accuracy of different models are verified and analyzed by using different forecasting methods and historical data, and the conclusion that the fitting effect of the modified multivariate regression prediction model is optimal is obtained. 4. In view of the optimal stage cycle regression forecasting model, three different investment operations are proposed to carry out portfolio empirical research, including investment based on the stage cycle regression prediction model. Do not follow recommendations for inaction and idealized indexed investments. The empirical results show that the portfolio of stock selection and investment operation by using stage cycle low and high point regression forecasting model is much higher than other models and can effectively avoid systemic risk. In the course of the operation of the forecasting model based on the stage cycle regression, the paper puts forward some countermeasures and suggestions for the stock selection and selling.
【學(xué)位授予單位】：天津理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2012
【分類號(hào)】：F832.51;F224

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 周萬(wàn)隆,吳艷;馬克維茨投資組合模型的遺傳算法[J];商業(yè)研究;2004年01期

2 吳世農(nóng),許年行,蔡海洪,陳衛(wèi)剛;股市泡沫的生成機(jī)理和度量[J];財(cái)經(jīng)科學(xué);2002年04期

3 徐愛(ài)農(nóng);;中國(guó)股票市場(chǎng)泡沫測(cè)度及其合理性研究[J];財(cái)經(jīng)理論與實(shí)踐;2007年01期

4 賈曉東,潘德惠,陶艷秋;證券投資組合理論在中國(guó)的適用性分析及其改進(jìn)[J];東北大學(xué)學(xué)報(bào);2002年08期

5 申隆;;上海證券市場(chǎng)投資組合規(guī)模、風(fēng)險(xiǎn)和收益關(guān)系的實(shí)證研究[J];電子科技大學(xué)學(xué)報(bào)(社科版);2006年06期

6 白],張世英;擴(kuò)展SV模型及其在深圳股票市場(chǎng)的應(yīng)用[J];系統(tǒng)工程;2001年06期

7 張永東,何榮天;深圳股市波動(dòng)性與成交量關(guān)系的實(shí)證分析[J];系統(tǒng)工程;2002年03期

8 胡海鵬,方兆本;用AR-EGARCH-M模型對(duì)中國(guó)股市波動(dòng)性的擬合分析[J];系統(tǒng)工程;2002年04期

9 王春峰,張慶翠;中國(guó)股市波動(dòng)性過(guò)程中的長(zhǎng)期記憶性實(shí)證研究[J];系統(tǒng)工程;2004年01期

10 林丹,李小明,王萍;用遺傳算法求解改進(jìn)的投資組合模型[J];系統(tǒng)工程;2005年08期

相關(guān)碩士學(xué)位論文 前1條

1 彭耿;投資組合理論及其績(jī)效評(píng)價(jià)模型的應(yīng)用研究[D];中南大學(xué);2003年

本文編號(hào)：2388547