【作者】 危慧惠；

【导师】 张宗成；

【作者基本信息】 华中科技大学 ， 西方经济学， 2004， 博士

【摘要】 可转换债券定价高低直接影响到发行人的融资成本和投资人的获利空间。我国的可转债市场近几年来发展迅速,选择一种适用于我国现状的可转债定价模型既是投融资者迫切的愿望,也是我国资本市场发展的急切需要。我国的可转债市场自从成立以来,已有许多对于可转换债券的定价研究。但是在研究方法上存在一些误区。主要有以下几个方面:第一,学术论文普遍套用国外的模型,虽然在理论上有一定的参考价值,但是不太尊重我国的实际情况;第二,各个证券机构的研究报告偏向于研究定性的转债条款,缺乏定量研究;第三,从已有的实证研究来看,可转债的选择比较单一,通常选定的是表现较好的几只转债,缺少对于整个转债市场的分析。另外样本数据跨越的时段比较短,一般是几天,最多也不超过一年, 因此不太具备说服力。为打破以上误区,本文进行了以下研究:首先介绍了可转债的概念,阐述了我国可转债的发行、上市及交易程序。通过与海外的可转债市场的比较,我们发现,我国可转债的缺陷在于我国资本市场的大环境欠佳,根本上还是制度原因造成的。接着本文给出了两种简单的可转债的定价方法。而对于精确的定价,我们则根据衍生证券的无套利定价原理,推导出可转债的价值方程进而根据数值算法进行求解。二项式算法是通常应用于期权定价和可转债定价的一种数值算法,因为该方法无需用到价值方程,简单准确而为业内人士广泛采用。本文就二项式算法的原理,精度以及几种版本的二项式方法等作了详细的介绍。在以上分析基础上,本文讨论了目前我国上市可转债的共性和特性。通过对可转债条款的比较分析,得出了一般性的投资策略。为精确定价,本文归纳和分析了我国可转债市场中采用的几种典型的模型,即简单组合模型、有限差分模型和二项式方法,并结合具体计算实例,对其定价方法的假设、原理及算法进行了比较。最后本文对我国可转债市场进行了实证研究。研究中采用2003年7月1日至2004年7月1日每周收盘数据,用组合模型和二项式模型对深沪股票交易所上市的22只可转债进行了理论定价。通过比较和讯网提供的理论值、组合模型理论值和修正二项式模型理论值三种可转债理论价值与实际市场观测价格之间的偏差,进而寻求可转债市场的内在规律。实证结果发现现阶段我国的可转债市场并不存在明显的套利机会,存在一定的风险;二项式方法产生的理论值最接近实际值,尤其对于实值可转债更是如此。总的来说,本文研究工作中的创新点有以下几个方面: <WP=5>第一,在样本和数据方面,通盘考虑了现阶段我国各类上市可转债的性质及特点。鉴于国内对可转债的实证研究多集中于个别优质可转债,而对整个可转债市场的分析较少的事实,本文选择了行业、流动性和所处时期各异的22只可转债作为研究对象,根据不同条款的可转债分析我国可转债市场的规律,进而确定正确投资策略。所得结论对于发行可转换的公司和投资可转债的机构具有一定的参考价值。第二,在模型的选择方面,采用了改进的二项式模型,主要是对于我国目前采用的波动率、折现率、利率以及二项式步长的选取方法做了修正。用修正后的模型对目前上市交易的22只可转债近一年的周数据进行了估计,并与目前机构采用的模型比较,结果发现修正后的模型能够更好的拟合实际观测价格。在模型的实现方面,没有采用国内业界人采用的Excel软件[70][71]。而是根据可转债条款的嵌入特性,针对对于输入数据的特点,采用了更为直观和实用的Matlab软件来实现定价过程。从而为更为复杂的可转债的定价打下了一个基础。第三,在对模型误差的分析上,比较了和讯网提供的理论值、组合模型理论值和二项式模型理论值的偏差程度,并对模型产生的价格与实际收盘价格的偏差产生的原因进行了分析。实证结果发现二项式结果的偏差最小,另外与和讯网给出的可转债价值被大幅低估的评价不同,二项式模型计算出来的结果表明现阶段我国可转债的价格总体来说不存在明显的低估或高估的,可转债市场存在一定的风险。本文对这种现象的原因做了分析,并进一步根据我国的实际,提出了消除这些偏差的几点建议。更多还原

【Abstract】 The pricing of Convertible Bonds(CB) is more and more important to the issuers and the investors. The overpricing or underpricing of CB will affect the financing cost to issuers and the profit of the investors. The CB maket of China is developing quickly nowadays. It is necessary to choose a suitable model to CB pricing of Chinese market .There have much research on the pricing of Convertible Bonds in China. But in my opinion, there are three main shortages of these papers. Firstly, the academically papers apply mechanically of the model of other developed countries, which is not suit to China. Secondly, most of the research reports of stockjobbers use qualitative analysis to determine the invest value of the Convertible Bonds. Thirdly, according to the recent empirical research, the CBs chosed to be studied is too single, lack of the study of the whole CB market of China. In addition, the range of the data is too short, just from several days to several months, so the result is not persuasive.Because of the above limitation, the paper introduces the concept of CB,how the CB is issued and the ways to trade in Chinese market. Comparing with the aboard CB market, we found that the limitation of CB market of China is brought by the macroecnomic environment. After that, the paper introduces two simple models to price the CB. In order to pricing CB accurately, we use the value equation of CB. Another common model is binominal model. Because of its simpleness and precision, binominal model is widely acceptable by the market participators.On basis of the general analysis of CB pricing, the paper then studies the commonness and the speciality of the trading CBs of Chinese market. Compared to the clauses of the CBs, we can determine the primary invest strategies. The paper gives the examples of how to realize the pricing programme by computer of the main models adopted by participators.The last and main part of the paper studies the CB market of China by collecting the actual data from July 2003 to July 2004. There are three model value of CB,the value supported by webside of Homeway, the value by component model and the value by binomial model. The result shows there is no evidence to prove the existing of arbitrage opportunity. The value by binomial model is the closest value to actual market value, especially to the In-the-money CBs.In one word, the main reseach in the paper are concluded as the following:Firstly, the paper considers the whole convertible bond market in China. Because the empiral research in China are mostly focused on one or two CB, but lack of the analysis of <WP=7>the whole market, the paper chooses the 22 different kinds of business and liquidity to be the objects of the study. Accoding to the different issues of the CBs, the paper gets the valuable invest stratege. The conclution is useful for the issuers and the investors.Secondly, the paper adopts the improved binomal model. The development is mainly the way to decide the volatility, the discount rate and the riskfree rate.Then studies the CB market of China by collecting the actual data of the recent one year by use the improved model. The result shows that the new model can fit the actual data better. On realizing the model, the paper chooses the more practiceful software, Matlb, instead of the Excel.Thirdly, the paper analysis the bias of the pricing by the comparsion of three different models. The result shows that the bias of the binomal model is smallest and the whole CB market in China is not be overpricing or underpricing. The result means that there are uncertain risk in the market of China. In addition, the paper gives some useful advice to reduce the bias.更多还原