【作者】 赵鹏；

【导师】 唐齐鸣；

【作者基本信息】 华中科技大学 ， 数量经济学， 2008， 博士

【摘要】 随着我国金融体制改革的深化和股票市场规模的扩大,股市中的泡沫成份日益增加。在资本市场开放度不断提高的情况下,股票市场泡沫的不断增加无疑将加大我国股票市场的风险,严重威胁我国金融体系的稳定。因此对我国股票市场泡沫的形成机制进行深入探究,对不同类型泡沫存在性问题进行实证检验,无疑具有重要的理论和实践意义。本文首先对于世界资本市场发展过程中的著名投机泡沫历史事件进行了回顾,对国内外已有的投机泡沫理论和实证文献进行了归纳和总结,并且提出了本文的研究方法和创新点。其次,分别对于股市理性和非理性投机泡沫形成机制进行了深入的阐述:在理性泡沫形成机制的阐述中,利用鞅模型推导出了股票内在价值和理性投机泡沫的定价模型,继而对理性投机泡沫进行了合理分类,并且具体分析了我国理性投机泡沫的具体形成机制;本文接着阐述了非理性投机泡沫模型,从股票市场信息不对称、投资者决策的非理性等方面对于股市非理性投机泡沫形成、膨胀和破灭的具体机制进行了深入的分析和探讨,具体包括“狂热和恐慌”模型、时尚泡沫模型、噪声交易者模型、正反馈交易者模型,以及群体行为的模仿和传染模型等,并且利用非理性投机泡沫理论对于我国股市非理性投机泡沫形成机制进行了具体的和深入的分析。再次,对于已有文献中股市投机泡沫的实证检验方法进行了具体的介绍,特别是对于检验的理论思路,以及检验方法的优势和弊端都进行了深入的分析。在此基础之上,本文分别采用了两种新的思路分别对于我国股市周期性破灭型投机泡沫和理性投机泡沫的存在性问题进行实证分析,这正是本文的创新之处:一方面,采用Van Norden和Schaller根据VNS模型(1993)提出基于指数超额收益率和相对泡沫度的动态关系的区制转换检验方法,此方法属于直接检验方法,理论背景较为深厚,且在实证研究中对泡沫现象的解释能力较强。基于这种方法,本文采用Markov区制转换模型对我国股市周期性破灭型泡沫的存在性进行了计量分析,得出沪深股市在2006-2007年间存在周期性破灭型投机泡沫的结论,并与MSVAR方法相结合,对股市正负泡沫存在的时期进行了区分。另一方面,将持续期依赖分析方法和状态转换模型相结合,使用具有持续性依赖特征的马尔可夫区制状态转换方法对于沪深股市的理性投机泡沫存在性进行了分析,这是一种间接检验方法,该方法的优势是,能够在识别理性泡沫存在时期的同时,深入揭示理性泡沫的持续性依赖特征,利用该方法得出沪深股市在2007年出现了理性泡沫的结论。最后,本文结合我国股票市场发展的现状和投资者的行为特点,具体分析了我国股市泡沫的成因,分别为:宏观经济长期的高速平稳增长、低通胀、本币持续升值是产生大规模股市泡沫的基础经济条件、优质股票供应不足导致股票供需失衡是泡沫产生的重要原因之一;大股东进行的内幕交易和价格操纵行为致使定价机制的扭曲,最终导致泡沫的急剧膨胀;个人投资者和部分机构投资者的投机心理非常严重,呈现出很强的正反馈交易和羊群行为特征,追涨杀跌现象十分普遍,导致股市呈现泡沫产生、膨胀、破灭的周期性循环的特征。本文基于泡沫的具体形成机制以及投机泡沫实证结果提出了相应的政策建议,具体包括股市适度扩容、加大对于内幕交易和价格操纵的监管力度、提高上市公司质量、规范政府行为、建立上市公司强制现金分红制度、完善市场套利机制、发展多层次的资本市场、加强投资者风险教育等,以便更好地防范和化解股市投机泡沫风险。更多还原

【Abstract】 With the finance system reform being deepened and the scale of sock markets being expanded, the bubbles in Chinese stock markets are increasingly augmented. In the background of capital markets being increasingly disparked, the bubbles in the stock market would amplify the risk of market crashes and thresten the stability of Chinese finance system. So It is of great importance to illustrate the generation machanism of speculative bubbles, and make empirical sudies on detection of bubbles in Chinese stock markets.Firstly, the paper narrated the famous events of speculative bubbles and summed up the financil theory and empirical papers, then put forward the study methods and innovation of the paper.Secondly, the paper revealed the generation machanism of rational speculative bubbles and irrational speculative bubbles. In the illustration of rational speculative bubbles, the paper utillized the martingle theory to induce the pricing equation of rational bubbles., then divided rational bubbles into several classfications. In the illustration of irrational speculative bubbles, the paper illstrated the generation machanism of irrational speculative bubbles from information asymmetry and irration of traders which include "panics and manias", "fads" , noise trader model, positive feedback model, and herd behavior model.Thirdly, the paper introduced the precent empiricl test models, particularly revealed the thoughts of the empiricl test models, then illustrate the advantage and defect of these models. On the base of above sudies the paper adpopted two innovative methods to detect thespeculative bubbles in Chinese stock markets:On one hand, the paper adopted regime switching test which were proposed by Van Norden and Schaller (1993) to detect the speculative bubbles in the stock markets. The regime switching test being a direct test method which has a profoundly theory basis and a high efficiency. The writer adopted markov switching model which based on regime switching test to detect the periodically collapsing speculative bubbles in Chinese stock markets, then concluded that there existed periodically collapsing speculative bubbles in Shanghai and Shenzhen stock markets in last two years. The paper linked regime switching method with the Msvar model to detected the positive bubbles and negative bubbles in Chinese stock markets.On the other hand, the writer linked duration denpendence test with markov regime switching test to acquire the duration denpendence markov regime switching test model (DDMS) ,which has great advantages over the other simple test methods, then utillized the DDMS model to test the rational bubbles in Chinese stock markets. The paper concluded that there existed rational speculative bubbles in Shanghai and Shenzhen stock markets in the last year.Lastly, the writer associated the precent condition of Chinese stock markets with the behavior feature of traders in Chinese to analysis the reason why speculative bubbles generated in Chinese stock market, then proposed several suggestions to prevent from the risk of speculative bubbles, which included to amplify the scale of stock markets, to strengthen the supervision of inside trade and price manipulation, to strengthen the hedge machanism in stock markets, and to make the traders more rational.更多还原