著者
石井 健一 ISHII Kenichi
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2011-08

マイクロブログとして世界で最も使われているTwitter のAPI を利用して発言を19,409件収集し、発言頻度やフォロー関係などを日本語利用者と英語利用者の間で比較した。その結果、次のような知見が得られた。(1) 日本語利用者は、フォロワー、フレンドの数は英語利用者よりも少ない。(2) 日本語利用者は発言頻度が多く、「お気に入り(like)」を使う頻度も多い。(3) 日本語利用者は、フォロー関係の相互性が高く、さらにフォロー数の多い人ほど相互性が高い傾向がある。こうした特徴は、日本人利用者の文化的特性を何らかの形で反映していると考えられ、利用者の動機など心理学的な視点からの比較研究が今後の研究課題となる。
著者
石井 健一
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2009-05

女性600 人(8-29 歳)を対象としたキャラクター消費に関するアンケート調査を分析した。キャラクターを好む動機の因子分析から、キャラクターを好む動機には「癒し」「対人動機」など6 因子があることを見出した。年齢別の比較では「癒し」は20 代、「対人関係」は10 代以下で高かった。キャラクター別では「トトロ」「くまのプーさん」「リラックマ」は「癒し」、「ポケモン」「スヌーピー」「ハローキティ」は「対人動機」の得点が高かった。商品カテゴリー別に、これらの動機との関連をみると、「癒し」が購買意図と関連している商品カテゴリーは「ぬいぐるみ」と「置時計・掛け時計」などがあったが、「対人動機」の方がより多くのキャラクター商品の購買意図と関連していた。つまり、キャラクターを好む動機とキャラクター商品を購買する動機は、必ずしも一致していなかった。
著者
石井 健一 ISHII Kenichi
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2012-01

本稿は、米国のランダムサンプリング調査(N=2252)に基づいてタブレット端末の普及要因を分析したものである。タブレット端末は、一般にはパソコンを代替するものだとみられているが、本分析結果でもノートパソコンの利用がタブレット端末利用と関係していることが示された。しかし、実際はタブレット端末の所有者は、パソコンの所有者とは大きく異なっていて、デモグラフィック要因との関係も異なっていた。また、コレスポンデンス分析の結果から、iPadと電子ブックリーダ(Kindleなど)は所有者にとって類似したものと位置づけられていた。タブレット端末とパソコンの間の代替関係については、将来の検討課題として残された。
著者
AKIYAMA Eizo HANAKI Nobuyuki ISHIKAWA Ryuichiro
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2013-08

To what extent is the observed mis-pricing in experimental asset markets caused by strategic uncertainty (SU) and by individual bounded rationality (IBR)? We address this question by comparing subjects initial price forecasts in two market environments – one with six human traders, and the other with one human and five computer traders. We find that both SU and IBR account equally for the median initial forecasts deviation from the fundamental values. The effect of SU is greater for subjects with a perfect score in the Cognitive Reflection Test, and it is not significant for those with low scores.
著者
YAMAMOTO Yoshitsugu ZHOU You
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2013-08

This paper is concerned about the aggregation function which plays a centralrole in the majority judgement that was recently proposed by Balinski and Laraki as anew voting mechanism. We raise two issues about their aggregation function, named orderfunction, and show that they are resolved by relaxing the strong monotonicity conditionimposed on the aggregation function, and that the anonymous, weakly monotonic andstrategy-proof aggregation function is completely determined by the set of final grades when the judges split deeply.
著者
IZUNAGA Yoichi YAMAMOTO Yoshitsugu
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2013-07

The modularity proposed by Newman and Girvan is the most commonly used measure when the nodes of a graph are grouped into communities consisting of tightly connectednodes. We formulate the modularity maximization problem as a set partitioning problem, and propose an algorithm for the problem based on the linear programming relaxation. We solve the dual of the linear programming relaxation by using a cutting plane method. To mediate the slow convergence that cutting plane methods usually suffer, we propose a method for finding and simultaneously adding multiple cutting planes which may complement well each other.
著者
OGAKU Michiko
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2013-06

This paper investigates why some business firms do not find it necessary to relyon explicit contracts to resolve incentive problems, even though they have recourseto contractible financial measures. Implicit incentives, such as career concerns,play a key role in those firms. Using the Normal-Exponential (CARA-Gaussian)model and assuming that the firms in this study have both contractible and noncontractible measures, this paper implies that some firms may wish to give their managers disincentives, if possible, because their managers have been overworking.
著者
OGAKU Michiko
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2013-06

This note provides the criterion for an information system for an optimal incentivescheme under plausible assumptions about firm managers: the agent has careerconcerns and there are not only contractible observables but also non-contractibleobservables. The covariance relation between the likelihood ratios of the informationsystems is found to be sufficient to recognise the information systems’ feasibility as a basis of an incentive scheme. Furthermore, it is shown that the criterion can specify more unfeasible information systems than Kim’s mean preserving spread criterion.
著者
OGAKU Michiko
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2013-05

This paper studies the economic consequences of choosing two differenttypes of executive compensation contracts. The analysis is based on a two-periodagency model in which compensation contracts are subject to renegotiation; compensationis paid based on the agent’s earnings report (e.g., a performance-based contract)or a non-verifiable measure within the firm (e.g., a conventional implicit contract).According to the analysis, conventional implicit contracts can dominate performancebasedcontracts if the non-verifiable measure is sufficiently informative so that theagent’s earnings report is not significantly considered during renegotiation. However,if the agent has strong bargaining power, the performance-based contract is optimal.The theoretical findings have implications for empirical compensation research. First,the firms’ compensation policy may not serve as a useful test for identifying profitablefirms. Second, the combination of the compensation policy and the ownership structureis likely to be associated with the level of executive compensation.
著者
IGARASHI Ayumi YAMAMOTO Yoshitsugu
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2012-11

In this paper we prove that calculating the average covering tree valuerecently proposed as a single-valued solution of graph games is #P-complete.
著者
KABE Satoshi KANAZAWA Yuichiro
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2013-02

In Bayesian data analysis, a deviance information criterion (DIC)proposed by Spiegelhalter et al. (2002) is widely used for the modelselection, since this criterion is relatively easy to calculate and applicableto a wide range of statistical models. Spiegelhalter et al. (2002)gave an asymptotic justification of DIC in the case where the numberof observations grows with respect to the number of parameters.In small-sample cases, however, the estimated asymptotic bias of DICmight underestimate the true bias (Burnham, 2002). In this paper, wepropose a finite-sample bias corrected information criterion (ICBL) forthe Bayesian linear regression models with conjugate priors, as AICCproposed by Sugiura (1978) in frequentist framework. We examine theperformance of the proposed information criterion relative to the DICfor small-sample cases by simulation, and found that our proposedinformation criterion outperforms DIC.
著者
MATSUKAWA Yasuaki YOSHISE Akiko
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2011-11

We call a positive semidefinite matrix whose elements are nonnegative a doubly nonnegativematrix, and the set of those matrices the doubly nonnegative cone (DNN cone). The DNN coneis not symmetric but can be represented as the projection of a symmetric cone embedded in ahigher dimension. In [16], the authors demonstrated the efficiency of the DNN relaxation usingthe symmetric cone representation of the DNN cone. They showed that the DNN relaxation givessignificantly tight bounds for a class of quadratic assignment problems, but the computational timeis not affordable as long as we employ the symmetric cone representation. They then suggested aprimal barrier function approach for solving the DNN optimization problem directly, instead of usingthe symmetric cone representation. However, most of existing studies on the primal barrier functionapproach assume the availability of a feasible interior point. This fact means that those studies arenot inextricably tied to the practical usage. Motivated by these observations, we propose a primalbarrier function Phase I algorithm for solving conic optimization problems over the closed convexcone K having the following properties: (a) K is not necessarily symmetric, (b) a self-concordantfunction f is defined over intK, and (c) its dual cone K∗ is not explicit or is intractable, all of whichare observed when K is the DNN cone. We analyze the algorithm and provide a sufficient conditionfor finite termination.
著者
Nagayasu Jun
出版者
University of Tsukuba. Graduate School of Systems and Information Engineering. Doctoral Program in Social Systems & Management
巻号頁・発行日
2006-11

We analyze the time-series properties of the dividend-price ratios (DPRs) of 11 developed countries since the early 70s. Despite its frequent use in research as a valuation method for stock prices and a determinant of stock returns, previous studies suggest that there is mixed evidence of the time-series properties of DPRs predicted by economic theory. We argue that this mixed result is attributable to the sample size used in previous studies. Here, we have opted to implement the panel data approach (i.e., N > 2) to increase the total size of observations rather than relying on the traditional method (i.e., increasing the size of T). In this way, we can increase the total number of observations without increasing the likelihood of structural breaks. For this purpose, we implement the panel unit root test taking account of cross sectional dependence,and obtain clear evidence of stationary DPRs. Thus, we conclude that a signi.cant one-to-one long-run relationship exists between stock prices and dividends for the countries in our samples.