【作者】 胡曙光；

【导师】 吴立扬；

【作者基本信息】 武汉理工大学 ， 金融工程与管理， 2007， 硕士

【摘要】 在经济发展过程中，交通基础设施起着决定性的作用，随着经济的发展，需要建设更多的高速公路满足日益增加的交通量。以净现值法为代表的传统投资决策方法由于存在忽视管理柔性和战略灵活性的缺陷，目前高速公路投资所采用的方法是沿用传统的投资决策方法，已越来越难以正确评价高速公路投资项目的真实价值，而实物期权法可以弥补这个不足，将项目中可能出现的各种机会、灵活性及不确定性通过动态复制当作期权进行一定程度的量化处理，从而提高复杂的投资决策的准确性。本论文尝试将实物期权理论引入高速公路项目投资评价之中，利用实物期权在发掘投资项目的弹性价值、管理不确定性、评估投资机会价值方面的天然优势，更贴近实际地评估高速公路项目的价值。但是由于传统B-S模型的参数难以确定，不符合高速公路投资决策的特点，所以本文结合BP神经网络以及遗传算法进行评价，用非解析解的形式，改进现有高速公路投资决策方法，吸引更多的投资者投资到高速公路投资领域，解决目前我国高速公路投资资金来源问题，同时也可以帮助决策者得出正确的决策分析。全文共分六章：第一章为导论，阐述了本论文的选题的目的和意义，与论文有关的国内外研究综述，以及研究内容和研究方法。第二章是提出问题，分析高速公路投资决策现状以及存在的问题。主要针对的是目前高速公路投资决策中所采用的传统NPV方法，并提出了基于实物期权的应用框架。第三章是根据应用框架，提出了高速公路的投资决策的基本理论，包括实物期权理论以及实物期权定价理论、BP人工神经网络理论、遗传算法理论。针对高速公路投资决策进行分析包括投资特点，影响因素，以及实物期权特性，为模型建立做准备。第四章是实物期权模型的建立，包括影响实物期权的因素分析，BP神经网络进行样本训练，运用遗传算法选择变量优化BP网络。第五章是具体的实证应用，用具体的高速公路投资项目来探讨基于B-P神经网络的实物期权的应用。第六章是结论部分，表明这种方法在高速公路投资决策比其它方法更科学、有效，并对应用前景进行展望。更多还原

【Abstract】 In the development of economy developing process, transportation infrastructure is playing an important function, with development of economic, there are more and more investments investing on highway built for traffic volume. And because of ignoring management and strategic flexibility, traditional investment decision-making method presented by net present value, has appraised the real value of highway investment project m ore and more difficulty, but real option may make up this insufficiency, to a certain degree, quantitatively processing each kind of opportunity, flexibility and uncertainty in the project by dynamic duplication, thus enhances complex investment decision-making accuracy.This thesis attempts to introduce real option theory to highway project investment appraisal, real option has natural superiority on excavating elastic value of investment project ,managing uncertainty, evaluating investment opportunity, but because parameters of traditional B-S model are hard to determine, this thesis combines BP nerve network with genetic algorithm to evaluate, with non- analytic solution form, improving decision-making method of existed highway investment decision-making methods , helping policy-maker make correct policy-making analysis.This thesis consists of six chapters:Chapter 1 is the preface, which introduces purpose and meaning of the field, the related research summarize of domestic and overseas, the purpose of research, content and method.Chapter 2 puts forward problems, analyses high way investment decision-making actuality and existed problems, mostly aiming at traditional NPV method, and putting forward applied frame based on real option.Chapter 3 put forwards elementary theory of highway investment decision-making according to applied frame, including real option and real option pricing theory、BP annual nerve network theory, genetic algorithm theory, makes preparation for establishing the model, analyzing investment decision-making including investment characteristic, influencing factor, as well real option characteristic.In the following chapter, the author puts forwards basal theory, including realoption theory and real option pricing theory、BP annual network theory、geneticalgorithm theory, and analyses investment characters of highway investment and its real option characters, preparing for model.Chapter 5, the author discussing on application of real option based on B-P nerve network by highway investment project.At the last chapter, the author draws the conclusion that the application of real option method is much effective and scientific than other methods.更多还原