Sociological Theory and Methods
Online ISSN : 1881-6495
Print ISSN : 0913-1442
ISSN-L : 0913-1442
Volume 14, Issue 2
Displaying 1-12 of 12 articles from this issue
Special Section
  • Kaoru ENDO, Tatsuhiro SHICHIJO
    1999 Volume 14 Issue 2 Pages 2_1-2_2
    Published: 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
    Download PDF (103K)
  • Formalization with Iterative Investment Game
    Hiroshi HAMADA
    1999 Volume 14 Issue 2 Pages 2_3-2_17
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         The purpose of this paper is to formalize a simple model that theoretically connects individuals' rational choice at micro level to income distribution, which is according to the Gibrat's law empirically, as social structure at macro level. We employ iterative investment game as a baseline model in which player has a binary choice between investing and not investing. Given parameters which prescribe payoff structure of the game are the prize density γ and the rate of return R . Method of analysis is a simulation with computation.
         We investigated changes in Gini coefficient and skewness of the gross profit distribution, as the parameters varied as follows : 0≦γ≦1, R={0.5,1,2,3}, and n (the number of times that the game is repeated) = 5 or 10. As a result of analysis, we derived the implication that Gini coefficient increases up to critical point, where γ≦1/(R+1), then decreases as prize density increases, where γ>1/(R+1).
         Furthermore, we show that our model, with cumulative effect, generates a lognormal distribution under condition that γ>1/(R+1).
    Download PDF (626K)
  • Yuhsuke Koyama
    1999 Volume 14 Issue 2 Pages 2_19-2_31
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         In this paper, we use a computer-simulation method and examine the hypothesis that consumers' purchasing heuristics amplifies the competing power of the low-involvement products. Under these two assumptions: (1) consumers use not optimizing but satisfycing and (2) shop managers follows POS-data and controls the exposing rate of products so that products with low competing power are immediately eliminated from the shop, the simulation results say these four results: (1) products with low competing power are immediately eliminated from the shop, (2) in case of highly competable products, there is a good circle that consumers' easily find the product which causes high sales which causes consumers' much easier finding and so on, (3) it is difficult to know in advance which product can survivve, (4) regardless of how severe competition among products, consumers can enjoy the stable consumer surplus from purchasing products.
    Download PDF (598K)
  • A Simulation Study
    Isamu OKADA, Kazunari ISHIDA, Toshizumi OHTA
    1999 Volume 14 Issue 2 Pages 2_33-2_52
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         According to the hypothesis of bounded rationality, there is no best approach for a reorganizing process due to the congnitive limit of organizational members. However, the reorganizing process can be successfully described in our model employing their attitude changes, in that the attitude change of personalistic agent causes a change of organizational behavioral pattern due to the attitude fluctuation.
         A multi-agent model based on a personalistic agent is developed to describe the reorganizing process in an organization with respect to an organizational performance and its environmental adaptation. A computer simulation is to employ to describe the reorganizing process, because the process must be difficult to deal with an empirical method due to the complexity. We also propose the approach as an operational organizational oriented approach, and discuss a methodological possibility of our approach.
         The attitude fluctuations in the reorganizing process are formulated in the model employing a concept of achievement motivation theory in psychology and of a differential engine proposed by Minsky. The personalistic agent consists of a task stickness, an emotion of interpersonal attractiveness, and of a conservativeness as variables.
         In conclusion, we may provide a mechanism of reorganizing and organizational rigidity as a mechanism of a positive and negative feedback loop mediated by the personality and the attitude fluctuations of organizational members.
    Download PDF (1017K)
  • Toshiyuki KANEDA
    1999 Volume 14 Issue 2 Pages 2_53-2_71
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         This paper explains simple mathematical models, all of which are devised by way of abstracting practical simulation models. First, following on Ackoff & Emery's discussion on ‘purposeful system’, the ‘conceptualization’ and ‘formalization’ of our ‘autonomy agent’ are presented. Each of the agent has an internal model, in order to deal with intellectual functions such as ‘cognition’, ‘learning’, ‘forecasting’, ‘planning’ and so on.
         The first model is formulated as a decision model under a bounded rationality caused by his/her internal cognition model. Some mathematical concepts are also defined such as three kinds of ‘learning processes’, on which his/her internal model are transformed. ‘Model space’ and other concepts are also introduced.
         The second model is formulated by introducing the first ones into a hierarchy system theory. This poly-agent system is designed as not a model of ‘incrementalism system’ but ‘hierarchical planning system’ for public administration. The hypergame theory is also referred as an example of game-theoretical situation consisted of the first ones. From the point of view of ‘problem solving- oriented’ modeling, a hybrid approach of such three kinds of model: mathematical model, computer multi-agent model, and gaming-simulation model, is presented. The mathematical models are expected to form a basis for designing the latter two operational simulation models; computer simulation and gaming simulation. Some potential implications on mathematical sociology are also suggested.
    Download PDF (1016K)
  • Yoshinori TOMIYAMA, Fumio HOSONO
    1999 Volume 14 Issue 2 Pages 2_73-2_88
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         “Laboratory assignment” is a process to assign students to laboratories for their graduation research. It assumes the following conditions: every student should belong to one and only one laboratory, the maximum number of students for each laboratory is fixed and announced in advance, and the process of assignment depends only on the preferences of the students and teachers involved. Under this assumption, we consider a laboratory assignment institution for each student to be able to belong to some laboratory. An extended version of the matching institution which was proposed by Gale and Shapley (1962) not only satisfies the conditions required but also has some theoretically desirable properties. But this institution assumes, among others, that every student should, by preference, linearly order all the laboratories from which to choose only one for his/her graduation research, which, though theoretically valid, is not realistically plausible. To put it into practice, it should be examined whether assignment can be successful when the number is relatively small of the candidates a student should choose first from all the laboratories. Analysis by Monte Carlo simulation reveals that this institution works well even in such situations, given a certain set of conditions on the maximum number of students for each laboratory and the number of the candidates the students should submit first.
    Download PDF (985K)
  • Constructing basic theories of simulations
    Tatsuhiro SHICHIJO,, Yasuto NAKANO,
    1999 Volume 14 Issue 2 Pages 2_89-2_101
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         This article intends to clarify properties of learning models in simulation studies and to review existing learning models. Learning models are often used in many simulation studies, but there is no uniform rule of learning. We introduce three technical characters (monotonicity, conditions of probability, neutrality) and three rationality characters (rationality in fixed situations, rationality in first order stochastic domination, rationality with risk preference in stochastic situations). We examine Michael Macy's model, the Erev & Roth model and some other models, and find that these models have different properties. Though learning is treated as one solution of social dilemma from the results of Macy's model (Kollock, 1998), Macy's model is a peculiar learning model. Learning is not always a solution of social dilemma. In this manner, a comparison of learning models from a uniform point of view clarifies the properties of each model and helps to study the conformity of a learning model and human behaviors.
    Download PDF (614K)
Articles
  • Taro KANEKO
    1999 Volume 14 Issue 2 Pages 2_103-2_107
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         This paper examines the meaning of neutrality theorem in the study of the logic of collective action and N-person Prisoners' Dilemma. Neutrality theorem implies perfect crowding-out. That is, if the government levies tax on contributors to finance more of the public good, then the government provision of public good will crowd out private contributions completely. And it can be shown that if the government collects some of the taxes from non-contributors, then the total supply of the public good will increase, but this policy is not Pareto-improving.
         These claims have three meanings in the study of the logic of collective action and N-person Prisoners' Dilemma. (1)Perfect crowding-out explains theoretically “the decline of voluntary cooperation” that Michael Taylor argued. That is, Taylor's argument can be shown without his assumption of preference changes in his own continuous strategy model.(2)These claims contradict earlier discussions in the theory of the state based on the logic of collective action or N-person Prisoners' Dilemma. Because these kinds of discussions assume the provision of public good by the state can correct the sub-optimality of voluntary contributions.(3)This argument sheds new light on “selective incentives”.
    Download PDF (685K)
  • Yasuyuki KIMURA, Kazunori YAMAGUCHI, Satoshi KITAYAMA
    1999 Volume 14 Issue 2 Pages 2_119-2_125
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         Connectivity is the index of social network analysis that deals with network data, and has been measured so far by tracing the adjacent relations between nodes. This algorithm is equivalent to the method called depth first search in the field of computer science. It is known that this algorithm is so efficient for detection of the depth of the data that has graph structure. Through taking advantage of Path theorem in graph theory and the nature of diameter of graph, however, it is possible to measure connectivity by simple matrix operation. In this study we present algorithm for measurement of connectivity based on matrix operation, and after demonstration of the algorithm using few examples we define connectivity as graph theoretical terms.
    Download PDF (300K)
  • Tsuneyuki NAMEKATA, Yoko NAMEKATA
    1999 Volume 14 Issue 2 Pages 2_127-2_133
    Published: September 30, 1999
    Released on J-STAGE: September 30, 2016
    JOURNAL FREE ACCESS
         In this paper we assume explicitly that a player has an incentive to a cooperative behavior in prisoners’ dilemma game and try to explain the realization of cooperative behavior. In addition to the usual utility-maximizer, we add some types of players who have a tendency to a cooperative behavior and show that some of the latter types contribute the realization of cooperation in a long period of interaction. Further in the replicator dynamics of the average payoff game it is shown that as the time goes to infinity the ratio of the selfish type decreases to 0% and that of Sticky Tit-for-Two-Tats (STF2T) increases. Also we propose a new learning model where the ratio of the selfish type remains the same and that of cooperative type increases in terms of STF2T.
    Download PDF (344K)
Book Reviews
Editor's Note
feedback
Top