双曲折现与时间一致投资决策

发布时间：2018-03-13 14:49

本文选题：双曲折现　切入点：偏好逆转　出处：《湖南大学》2014年博士论文　论文类型：学位论文

【摘要】：时间折现是目前行为经济学的前沿问题,其中如何选择一个合适的折现率在项目的成本收益分析中至关重要,尤其当项目涉及时间很长时,折现率的一个微小变化都有可能决定该项目能否通过成本-收益分析。从八十年代开始的实验经济学表明决策者的折现行为不满足新古典经济学通常的假设-指数折现。在各个研究领域,指数折现理论正逐渐被折现率递减的双曲折现理论替代。然而,一方面,从现实问题和实验结果脱胎而来的双曲折现理论受到了各种各样的质疑；另一方面,双曲折现导致最优决策时间不一致,存在不可执行性。而在连续随机问题中,由于问题的复杂性,很难求出具体的解析解。因此,研究双曲折现的合理性以及非指数折现下的时间一致投资决策问题具有重要意义。本文以行为个体为主要研究对象,第2、3和4章研究第一问题,第5章研究第二个问题。偏好逆转现象作为指数折现的‘异象’存在,常常被用来当作双曲折现合理性的证据。本文第2章分析偏好逆转现象是否总可以当作双曲折现合理性的证据。本章首先指出共时性和历时性偏好逆转现象的区别,并指出在折现函数满足平稳性假设时,二者等价；然后说明,当满足平稳性假设时,双曲折现能够解释这两种不同的偏好逆转现象；其次,将决策者的不确定性生命和遗产效用纳入到决策过程,构建了一个折现函数,在一定的情况下,该折现函数既不满足平稳性假设又不是双曲型,能解释历时性偏好逆转现象但是不能解释共时性偏好逆转现象,偏好逆转现象在一定的情况下不能够作为双曲折现合理性的证据。现有文献认为次可加折现现象不能被双曲折现模型解释,因而质疑双曲折现的合理性,第3章回答该质疑。本章首先指出,在进行跨期选择时,具有双曲折现偏好的决策者对同一区间的折现会随参考点的变化而变化,因此决策者可能希望以折现区间起点为参考点,决策者对参考点的选择会影响折现行为。有一些决策者倾向于低估未来偏好变化的程度-也称为投射偏见；然后指出这种投射偏好导致时间偏好被扭曲,这种扭曲可以看成是对次可加折现现象的一个可能的理论解释,并指出那些质疑双曲折现的文献忽略了参考点选择以及投射偏见对折现的影响,次可加折现现象不能否定双曲折现；最后指出,当采用第2章提出的折现函数,且将决策者按成熟程度分类,成熟型决策者和部分成熟型决策者也会具有次可加折现现象,决策者的参考点选择以及成熟性导致出现次可加折现现象。第4章利用拉姆齐最优增长模型,分析当资本产出率不确定时,确定性等价瞬时折现率的结构。首先得到确定性等价瞬时折现率等于在风险调整密度函数下折现率的期望值,其中风险调整密度函数由资本产出率的密度函数,延迟时间以及相对厌恶系数决定；然后指出确定性等价瞬时折现率随延迟时间的增加而减小,具有双曲性；最后给出了一个数值分析。第5章研究当时间偏好由随机准双曲折现描述时,Merton(1969,1971)提出的经典有限期限投资-消费决策的时间一致投资决策问题。首先给出投资消费模型以及决策者的时间偏好；然后分析有限期限成熟型决策者的HJB方程,并进而推出无限期限成熟型决策者的HJB方程；其次分析对数和幂效用函数下的时间一致消费-储蓄和投资组合策略；最后给出比较动态行为分析和数值分析。
[Abstract]:Time discounting is the current frontier issue of behavioral economics, including how to choose a suitable discount rate is very important in the cost-benefit analysis of the project, especially when the project involves a long time, a small changes in the discount rate are likely to decide whether the project through the cost-benefit analysis. From the experimental economics began in 80s show that the discount behavior of decision makers do not satisfy the assumptions of neoclassical economics usually discounted the index. In various research fields, the discount index theory is gradually being the discount rate decreasing hyperbolic discounting theory substitution. However, on one side, from the practical problems and the experimental results born out of the hyperbolic discounting theory by the various questions; on the other hand, hyperbolic discounting leads to the optimal decision-making time is not the same, there is no execution. But in continuous random problems, because of the complexity of the problem, it is not easy to produce Therefore, it is of great importance to study the rationality of hyperbolic discounting and the time consistent investment decision making problem under non exponential discounting. In this paper, behavior subjects are selected as the main research objects, the first and the 4 chapters are the first ones. The fifth chapter studies second problems.
The preference reversal phenomenon as index discount "vision exists, is often used as the rationality of hyperbolic discounting evidence. The second chapter analyzes the preference reversal phenomenon is always as hyperbolic discounting rationality of evidence. This chapter points out the difference of preference reversal phenomenon and diachronic synchronic, and pointed out that the discount function satisfies stationarity the two hypothesis, then, when the equivalent; meet the stationarity assumption, hyperbolic discounting can explain the two different preference reversal; secondly, decision makers will be included in the uncertainty of life and heritage to utility decision-making process, construct a discount function, under certain circumstances, the discount function is do not satisfy the stationarity assumption is not hyperbolic, can explain the diachronic preference reversal but cannot explain the preference reversal phenomenon is, the preference reversal phenomenon not in certain circumstances Sufficient evidence for the reasonableness of hyperbolic discounting.
The existing literature that can not be explained with the discount phenomenon of hyperbolic discounting model, thus questioning the rationality of hyperbolic discounting, the third chapter to answer the questions. This chapter points out in intertemporal choice, with hyperbolic discounting preferences of the decision makers on the same interval discount will change according to the reference point and therefore decision may wish to discount the interval starting point as a reference point, decision makers in the choice of the reference point will affect the discount behavior. There are some decision makers tend to underestimate the extent of future preference change - also known as projection bias; then pointed out that this projection to partial time preference is distorted, this distortion can be seen as a possible explanation for the the theory of subadditive discount phenomenon, and pointed out that those who questioned the hyperbolic discounting literature ignores effects of reference point selection and projection bias on the discount, can not add the discount phenomenon of double negation Finally, twists and turns; when the second chapter proposed the discount function, and making classification according to the degree of maturity, mature and mature type of decision makers decision makers will have a subadditive discounting phenomenon, the selection of reference points for policymakers and the maturity of the resulting sub additive discount phenomenon.
The fourth chapter is the analysis of the Ramsay optimal growth model, when the rate of the capital output uncertainty, certainty equivalent instantaneous discount rate structure. Firstly, get the certainty equivalent instantaneous risk adjusted discount rate equal to the density function under the discount rate expectations, the risk adjusted capital output ratio of the density function by density function, delay time and relative aversion coefficient the decision; and pointed out the certainty equivalent instantaneous discount rate increased with the increase of delay time decreases, the hyperbolic; finally a numerical analysis.
The fifth chapter studies when time preference is described by stochastic quasi hyperbolic discounting, Merton (19691971) classic finite period of investment and consumption decision made by the same time investment decision problem. Firstly, investment consumption model and the decision maker's time preference; HJB equation and analysis of finite period of mature type of decision makers, and then launched the HJB equation infinite period mature type of decision makers; followed by analysis of logarithmic and power utility function of the time consumption and savings and portfolio strategy; finally, compare the dynamic behavior analysis and numerical analysis.

【学位授予单位】：湖南大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：F830.59;F224

【参考文献】