考虑风险规避的秘书问题求解的一个启发式方法

秘书问题是一类序贯观察与选择问题,描述了动态的信息搜索与决策过程.针对现有的以寻找满意解为目标的启发式方法存在诸多局限,提出了新的启发式方法,该方法基于当前观测中侯选项在已观察侯选项中的相对排名、待观测侯选项数量以及决策者的抱负水平,决策者可以通过设定抱负水平灵活决定该启发式方法的结果导向.推导了该启发式方法的性能指标,并通过仿真的方法与已有启发式方法的性能进行了比较.结果发现,该启发式方法在最终选择的侯选项的期望排名和稳定性,以及风险解的避免上均优于已有的启发式方法.     

动态决策过程中备选项的序列特征效应研究

动态决策过程中备选项随机出现的序列特征,可以分为较优选项连续出现、间隔出现以及较劣选项连续出现等特征.运用实验室实验的方法研究发现,当较优选项连续出现时会"启发"决策者产生"手热"认知心理偏差,进而导致较优选项连续出现的次数与决策者的搜索数量呈正相关关系.而当较劣选项连续出现时则会"启发"决策者产生"赌徒谬误"认知心理偏差,进而导致较劣选项连续出现的次数与决策者的搜索数量也呈正相关关系.这解释了在动态决策过程中,决策者面对较优或较劣选项连续出现的情境下,为何都采取继续观望等待的非理性决策行为.同时,对"手热效应"与"赌徒谬误"效应的差异性以及出现的情境进行了比较分析.     

董事会里的"战争"——序贯与惩罚机制下董事会决策行为的实验分析

本文利用实验研究方法首次在独立董事人数占优的董事会中引入序贯和惩罚机制,探讨了董事会科学决策的促成因素及制度环境.实验结果发现,序贯与惩罚机制引入后,董事决策正确率较静态实验分别提高了39.09%和34.26%,但两者的作用原理不同:序贯机制着力于改善董事会的私人信息结构,通过将独立董事"知情化",从而使决策行为独立于董事类型;惩罚机制则是利用独立董事的社会偏好进行治理,使决策行为与项目类型无关.与Gillette等人的研究结果不同,好项目的通过率并不是100%,原因可能与序贯时的针锋相对策略、惩罚时的社会偏好存在有关.     

一类序贯决策方法的提出:透镜与直觉模糊信息的视角

与大多数关注经典秘书问题的序贯决策研究不同,本文针对决策对象不变而其状态更新的一类序贯决策问题,提供决策方法。相较已有研究而言,本文从透镜与直觉模糊信息的视角提出这类序贯决策的一般方法,更加具有普适性,可应用于应急管理、风险投资等多个方向。该决策方法具有如下特点：1）借鉴透镜模型,通过当期可观测的线索,预测这类序贯决策问题的下一期状态;2）考虑到决策环境具有不确定性,采用直觉模糊数对决策者获取的信息线索进行刻画;3）为最大化模拟现实决策者,在前景理论框架下提出状态更新的序贯决策方法。本文提出的决策方法依据在于决策者对决策对象的预测"前景"值和预测绩效,基于此提出了具体的决策步骤,并辅以案例进行分析。通过对不同决策情形下决策结果的比较分析和对案例的稳健性检验,验证了该决策方法的适用性和有效性。     

供应链中序贯信念修正的Bayes博弈模型

提出了使用序贯信念修正方法来削弱供应链中的信息不对称现象.假设零售商拥有关于市场需求的私有信息,供应商只能粗略了解零售商的最优决策方式,然后使用可观察结果的多阶段Bayes博弈模型来描述供应链的运作过程,在各阶段之间根据Bayes法则修正供应商的信念.分析表明序贯信息修正方法能够使供应商的信息依概率收敛到零售商的私有信息,信息不对称博弈也将依概率收敛到信息对称博弈,并且收敛结果与初始信念无关.数值模拟实验进一步验证了序贯信念修正方法的有效性.     

信息瀑布中的动物精神

信息瀑布行为已被研究者逐渐接受并作为经典金融市场理论的有力补充,但现有"一次决策"信息瀑布实验很难刻画这种信念依赖现象;赋予决策者对同一事件一次以上序贯选择机会的多决策信息瀑布实验更符合实际.结果表明:第二次决策阶段信息瀑布发生率显著提高,但决策正确率却有所下降,这为凯恩斯的"动物精神说"提供了实验证据.此外,平衡状态下位于决策序位中部的决策者系统性地背叛了自己的私人信息.     

网络视角的风险与时尚对消费者信息搜索行为影响研究

消费者决策一直是消费者行为研究的核心,信息搜索作为消费者决策的第一步受到理论和实践界的普遍关注。互联网时代,消费环境快速变化呈现出网络化特点,消费者追求时尚与个性化,产品技术复杂化增加了消费者购买的风险,这些因素使得消费行为越发嵌入于网络当中。基于此,文章从网络视角出发,以感知风险和时尚为切入点研究社会网络对消费者信息搜索行为的影响。在网络构建上,与通常直接建立网络的方法不同,作者将网络结构指标与回归分析相结合,通过对以往文献的梳理,构建了以网络规模、网络结构洞、网络凝聚子群和关系强度为指标的社会网络结构量表,采用回归的方法揭示社会网络对消费者信息搜索的作用。通过对样本消费者调查研究发现,时尚和社会网络影响消费者信息搜索行为。其中,时尚敏感度与消费者信息搜索行为之间存在正向关联。社会网络中的网络规模和关系强度会分别发挥正向和负向的调节作用,调节时尚敏感度与信息搜索行为之间的关系。该研究在一方面为今后网络角度的消费者行为研究在网络构建上提供帮助,另一方面验证了企业在进行产品设计、品牌营销时,可以借助社会网络,提高战略的针对性和有效性。     

基于潜在竞争者的公司信息披露策略研究——实验室实验的检验

潜在竞争者是公司重要的利益相关者,本文围绕公司的信息披露,构建“在位者—股东—潜在竞争者”的博弈模型,求解在位者与潜在竞争者交互行为的均衡结果,在此基础上进行实验检验,发现潜在竞争者有低估在位者信息披露程度的先验信念,行为决策表现为:当披露收益大于其进入成本时大都选择进入;当披露收益小于或等于进入成本时,其决策依据为披露收益与进入成本之差,当披露收益与进入成本之差较小时,一般会选择进入.研究还发现,在位者信息披露的决策信念是既要照顾股东的利益又要设法阻止潜在竞争者的进入,其行为决策表现为:当预期收益小于或等于进入成本时,执行较高程度的披露策略;当预期收益大于进入成本时,更多地选择较低程度的信息披露.另外不进入价值较大时,在位者也倾向于更低程度的信息披露.     

非正式层级对董事会决策过程的影响:政治行为还是程序理性

本文探索性地研究了转型环境中董事间非正式关系影响董事会战略决策的过程机制。利用与中国上市公司协会合作调查获取的上市公司董事会决策过程相关数据并结合上市公司数据库数据,考察了中国文化背景下,董事会非正式层级对决策过程的影响机理。研究发现,董事会非正式层级会增加决策过程中的政治行为、降低程序理性。在对政治行为和程序理性进一步细分后发现,董事会非正式层级通过提高董事的自我意识、个体影响从而增加政治行为,通过降低决策过程中信息搜寻、信息加工而降低程序理性。本文的研究揭示了转型环境中董事会非正式层级所具有的分化、竞争的功能,在一定程度上打开了董事会决策过程的"黑箱",丰富了董事会决策过程研究,为转型环境下公司治理行为研究做出贡献,亦对中国公司治理实践具有启示作用。     

考虑信息搜索的环境污染群体性事件最优决策模型

坚决打好污染防治攻坚战是建设全面小康社会三大任务之一,环境污染群体性事件影响社会稳定和生态文明建设。在当今信息渠道多元化的自媒体时代下,群众能够通过信息搜索行为为决策获取信息。本文考虑了群众信息搜索行为,构建周边居民信息搜索的环境污染群体性事件最优决策模型,分析了不同条件下均衡策略选择,以及在各种地方政府补贴水平和暗箱操作程度下,周边居民的收益变化情况和策略选择差异。结果表明：当政府的暗箱操作程度较低时,群众的最优化策略为信息搜索后“污染轻微”则“抗议”,否则群众的最优化策略为直接抗议。当群众对项目污染程度为严重的先验概率较低且政府的暗箱操作程度比较高时,群众将选择不抗议策略。当周边居民的识别能力较差,群众将会选择策略信息搜索后“污染轻微”则“不抗议”,当暗箱操作程度足够大时,群众将会选择策略直接抗议。随着地方政府提供补贴水平的增加,群众的抗议强度减弱。了解群众的最优决策,有利于政府对环境群体性事件进行防范和应对,维护社会稳定。     

Entropy Principles in Decision Making Under Risk

This paper considers decision problems where: (1) The exact probability distribution over the states of nature is not precisely known, but certain prior information is available about the possibilities of these outcomes; (2) A prior distribution over the states of nature is known, but new constraint information about the probabilities becomes available. The maximum entropy principle asserts that the probability distribution with maximum entropy, satisfying the prior knowledge, should be used in the decision problem. The minimum cross-entropy principle says that the posterior distribution is the one which minimizes cross-entropy, subject to the new constraint information. The entropy principles have not gone uncriticized, and this literature, together with that justifying the principles, is surveyed. Both principles are illustrated in a number of situations where the distribution is either discrete or continuous. The discrete distribution case with prior interval estimates based on expert opinions is considered in detail.     

REPORT FREQUENCY AND MANAGEMENT DECISIONS

There is no theory to prescribe how frequently accounting reports should be issued for management decisions; indeed, one research study could not find evidence that frequency of reports is a variable in the decision process. Accounting reports are discrete aggregations from continuous processes, and frequency affects information content of reports. If frequency affects information, it therefore must be a variable in the decision process. This research investigated the question of whether the report frequency had any effect on evaluations made from accounting reports issued on different intervals. Evidence is presented to support the conclusion that frequency is a factor in the decision process and that decision behavior is affected by the frequency of reports. This report merely demonstrates the existence of the report frequency variable; much more research is necessary to control the variable for optimal system design.     

A rolling test of granger causality between the Finnish and Japanese security markets

In the paper we test the impact of the Japanese stock market on two financial asset groups, free and restricted shares, on the Finnish market in the early 90s. The causality is tested in the Granger sense. The research issue is particularly interesting, since the restrictions on foreign ownership were abolished by the end of 1992. The linkage between the Japanese and Finnish financial economies is seen to be stronger for free shares than for restricted. In particular, significant Granger causality between Japanese and Finnish free shares is observed at relatively long consecutive time intervals, whereas the Japanese impact on the restricted shares is only occasional. Thus, the decision to abolish the restrictions not only leads to increased international dependence in the future, but will also change the risk profile of the restricted shares.     

Obtaining Informationally Consistent Decisions When Computing Costs with Limited Information

We demonstrate the need to view in a dynamic context any decision based on limited information. We focus on the use of product costs in selecting the product portfolio. We show how ex post data regarding the actual costs from implementing the decision leads to updating of product cost estimates and potentially trigger a revision of the initial decision. We model this updating process as a discrete dynamical system (DDS). We define a decision as informationally consistent if it is a fixed‐point solution to the DDS. We employ numerical analysis to characterize the existence and properties of such solutions. We find that fixed points are rare, but that simple heuristics find them often and quickly. We demonstrate the usefulness and robustness of our methodology by examining the interaction of limited information with multiple decision rules (heuristics) and problem features (size of product portfolio, profitability of product markets). We discuss implications for research on cost systems.     

基于CI-TOPSIS的梯形直觉模糊多准则群决策分类方法

模糊群决策分类方法广泛应用于政治、经济与社会生活各个领域,可有效避免个人知识与经验局限性所导致的决策失误。针对信息不完备的多准则群决策问题,提出基于CI-TOPSIS的梯形直觉模糊多准则群决策分类方法。首先,给出梯形直觉模糊集及广义梯形直觉模糊几何聚类算子,兼顾考虑群决策中相应依赖属性与决策者的决策偏好。其次,给出基于离散Choquet积分的TOPSIS算子（CI-TOPSIS）,以此为基础,进一步给出基于CI-TOPSIS的梯形直觉模糊多准则群决策分类步骤,用于确定具有最大可信度群体一致案例比较信息集,并逐步引导决策者给出部分及全部方案的精确分类,充分考虑模糊决策环境下决策者偏好与案例比较信息的级别关系。最后,通过一个投资决策实例对所提出的多准则分类方法进行验证。实例分析表明：该方法克服了决策过程中信息的遗漏,充分保留了决策过程中信息的完备性,更适用于直觉模糊群决策环境下的决策实践,是一种非常有效和科学的方法,可应用推广到更多决策领域。本文所得结论,对于有效解决多人多投资方案的群决策问题,具有一定的借鉴意义。     

Two ways to handle dependent uncertainties in multi-criteria decision problems

We consider multi-criteria group decision-making problems, where the decision makers (DMs) want to identify their most preferred alternative(s) based on uncertain or inaccurate criteria measurements. In many real-life problems the uncertainties may be dependent. In this paper, we focus on multicriteria decision-making (MCDM) problems where the criteria and their uncertainties are computed using a stochastic simulation model. The model is based on decision variables and stochastic parameters with given distributions. The simulation model determines for the criteria a joint probability distribution, which quantifies the uncertainties and their dependencies. We present and compare two methods for treating the uncertainty and dependency information within the SMAA-2 multi-criteria decision aid method. The first method applies directly the discrete sample generated by the simulation model. The second method is based on using a multivariate Gaussian distribution. We demonstrate the methods using a decision support model for a retailer operating in the deregulated European electricity market.     

沪市认购权证与其标的股票价格走势的Granger因果检验

运用G ranger因果检验的方法,检验沪市6只认购权证与其标的股票之间的因果关系。分析结果表明:在大样本条件下,认购权证与其标的股票之间存在单向的G ranger因果关系,权证的价格走势影响其标的股票的价格走势,说明从长期来看,认购权证发挥了其潜在的投资杠杆的作用,其价值发现与风险对冲功能在市场中得到实现;在小样本条件下,认购权证与其标的股票之间不存在显著的G ranger因果关系,表明短期内权证与其标的股票价格走势相互独立,市场中有可能存在对权证投机炒作,人为放大权证的投资风险,使权证背离其正常的投资价值。     

A COMPARISON OF INFORMATION STRUCTURES FOR A RIGID DECISION RULE CASE

The information economics model of individual decision assumes that decision makers are capable of instantly modifying their decision rules in response to signal changes from the information structure. Individuals, however, often tend to stick to rigid decision rules, particularly when they are trained to respond with conditioned reactions, such as to a traffic light. This article modifies the individual decision model for such cases. It discusses probabilistic information structures as well as the deterministic information function. It results in an ordering of information structures that differs from the ordering imposed by the Blackwell Theorem. When deterministic information functions are considered, an ordering that is finer than the prevailing one is proposed. Transformations between deterministic and probabilistic information functions and the relationship between the proposed model and the previous theory are also discussed.     

A Simulation of Partial Information Use in Decision Making: Implications for DSS Design

Information matrices are often the output produced by a decision support system. These matrices are a common method for expressing a decision situation under different decision-making scenarios. The decisions involved in designing a decision support system to generate the information matrix are important and involve several cost and benefit components. A designer needs guidance in making effective design decisions in this context. Such guidance can be provided by considering the relationships among specific design decisions, costs, and benefits. The general objective of this study is to provide a comprehensive framework for this purpose. This study is the first to develop and present a comprehensive cost-benefit framework for evaluating design decisions for a variety of scenarios. The specific objective of this research is to provide guidance regarding the number of available information dimensions to incorporate in a computer-based decision aid. Simulation experiments are conducted with a completely specified model based on the cost-benefit framework (including needed assumptions) to evaluate how many information dimensions to include for a specific information matrix size to achieve a balance between information use costs and decision quality. Based upon extensive simulation analyses for a hypothetical decision maker, the practical guideline found for designers is to include only the top half of the relevant information dimensions in any specific decision support system. Over a large number of repeating choice decisions, the savings in cognitive effort and information gathering costs clearly offset relatively minor losses in decision quality.     

An approach to consensus measurement of linguistic preference relations in multi-attribute group decision making and application

This paper studies the group decision making problem with linguistic preference relations. We first study the consensus measure between the individual preference relations and the collective (group) preference relation by defining the concept of degree of similarity between two linguistic values and two linguistic preference relations. Then we propose a concept of the acceptance consensus threshold value for group decision making with linguistic preference information. We show that the consensus between individual preference relations and the collective (group) preference relation is greater than the weighted similarity degree of a given individual preference relation with respect to other individual preference relations in group decision making with linguistic preference relations. The results will help in the analysis of crucial issues of conflict and agreement among preferences of decision-makers, which affect the consensus of group decision making with linguistic preference relations. Theoretical foundations are then established for the proposed method. Finally, the proposed method is applied to evaluate the degree of consensus of individual overall preference values with respect to the collective overall preference values for multi-attribute group decision making with linguistic information. The main contribution of this paper is twofold. One is to present a new way to measure the consensus between the individual preference relations and the collective (group) preference relation in group decision making with linguistic preference information. Another is to provide an effective approach to evaluating individual overall preference values with respect to the collective overall preference values in multi-attribute group decision making with linguistic information.