By how much does the presence of behavioral uncertainty in an experimental asset market reduce subjects’ confidence in their price forecasts? An incentivized interval forecast elicitation method is employed to answer this question. Each market consists of six traders, and the value of dividends is known. Two treatments are considered: six human traders (6H), and one human interacting with five computer traders whose behavior is known (1H5C). We find that while the deviation of the initial price forecasts from fundamental value is smaller in the 1H5C treatment than in the 6H treatment, albeit not statistically significantly, the average confidence regarding the forecasts is not. We further analyze the relationships between subjects’ confidence in their forecasts and their trading behavior, as well as their trading performance, in the 6H treatment. While subjects’ high confidence in their short-term forecasts shows a negative correlation with their trading performance, high confidence in their long-term forecasts shows a positive correlation with trading performance.

We conducted asset market experiments where one experienced subject interacts with five inexperienced subjects to investigate how experienced subjects change their price forecasts and trading behavior when faced with strategic uncertainty caused by inflows of inexperienced subjects. Only half the experienced subjects initially forecasted prices deviating more from the fundamental values in the market with five inexperienced subjects than in the final round of the experiment in which they had previously participated. Furthermore, the majority of our experienced subjects did not change their trading behavior. Many experienced subjects act as price stabilizers when the inflow of inexperienced subjects is not associated with other changes in market conditions. (C) 2014 Elsevier B.V. All rights reserved.

Exploiting unique panel data that include direct measurements of resource allocation within households, we investigated the impact of childbirth on intrahousehold allocation for married Japanese couples. Based on a collective model of the household, we developed reduced-form and structural-form estimation equations that allow us to focus on private goods to track the changes in intrahousehold resource allocation. We found one additional child is associated with a reduction in the wife's private expenditure share by at least two percentage points. This may be because she substitutes more say in decisions on the children for her own private expenditure share.

This paper shows that it is possible to track the changes in the distribution of power within a couple by focusing on the changes in the pattern of private consumption when the consumption decisions are efficient and private consumption is separable from public consumption in individual preferences. We first show that the separability of private consumption from public consumption at the individual level carries over to the household level. Hence, changes in public consumption only matters through a change in the residual budget available for private consumption. When the consumption decisions within the household is efficient, private consumption decisions can be modeled as the solution of a problem consisting in maximizing a weighted sum of the private-consumption sub-utility functions of the spouses under the residual budget, the weights being unique and representing the distribution of power over the allocation of private consumption. The model presented in this paper can be used to analyze the changes in the household resource allocation due to, for example, childbirth.

We consider a simple two-period model of irreversible investment under strategic interactions between two players. In this setup, we show that the quasi-option value may cause some conceptual difficulties. In case of asymmetric information, decentralized investment decisions fail to induce first-best allocations. Therefore a regulator may not be able to exercise the option to delay the decision to develop. We also show that information-induced inefficiency may arise in a game situation and that under certain assumptions inefficiency can be eliminated by sending asymmetric information to the players, even when the regulator faces informational constraints. Our model is potentially applicable to various global environmental problems. (C) 2011 Elsevier B.V. All rights reserved.

This paper presents a model of learning about a game. Players initially have little knowledge about the game. Through playing the same game repeatedly, each player not only learns which action to choose but also constructs a personal view of the game. The model is studied using a hybrid payoff matrix of the prisoner's dilemma and coordination games. Results of computer simulations show that (1) when all the players are slow at learning the game, they have only a partial understanding of the game, but might enjoy higher payoffs than in cases with full or no understanding of the game; (2) when one player is quick in learning the game, that player obtains a higher payoff than the others. However, all can receive lower payoffs than in the case in which all players are slow learners. (C) 2009 Elsevier B.V. All rights reserved.

We reconsider the no trade theorem in an exchange economy where the traders have non-partition information from aspect of rough sets. By introducing a new concept, rationality of expectations, we show some versions of the theorem different from previous works, such as Geanakoplos (1989, Came theory without partitions, and applications to speculation and consensus, Cowles Foundation Discussion Paper No. 914).

This paper presents a communication protocol to reach consensus. In our setting, every player has asymmetric information and evaluates a fixed event on his information. According to the protocol, the sender sends non-partitional messages that he believes the event with a probability of at least his evaluation. We show that the posteriors for the event must be equal among the players after the communication.

This article investigates epistemic conditions for a sequential equilibrium in an extensive form game with imperfect information: If players mutually know that all players maximize own expected payoffs for any information sets in their final decisions, then their behaviors with belief yield a sequential equilibrium. This result is an extension of [2] in a perfect information game. In this article, we propose the notion of mu-rationality, by which we mean that player knows that he maximizes his own payoff according to the common-belief mu. We also introduce the notion of mu-consistency in imperfect information game. Our main theorem states that mutual knowledge of both mu-rationality and mu-consistency induces a sequential equilibrium outcome in an extensive form game.

[金子守]制度・認識の問題を限定合理性の立場から考察した。特に、期待効用理論において確率の使用を限定する事を研究した。それによって、アレーのパラドックスのより良い理解とパラドックスからの回避を議論し、制度・認識のより良い理解が得られた。また、船木氏、竹内氏と実験の立場から帰納的ゲーム理論を考察した。[船木由喜彦]実験研究、協力ゲーム研究共に進展し、新しい知見が得られた。国際学会、国内学会でそれらの研究成果を報告した。[鈴木信行]ゲーム理論的な状況における意思決定過程の論理学的分析に関して、構成的推論の観点から研究した。omniscience principlesを中間述語論理の枠組みで研究し、omniscience principlesと中間述語論理の関係について、一定の結果を得た。[秋山英三]主体間の繋がりに濃淡があるネットワーク上において囚人ジレンマがプレイされる系に関して分析を行い、小規模な集団でLink-weightの異質性が協力進化に促すための条件を導出し、大規模集団でもその条件がある程度成立することを計算機シミュレーションで示した。市場取引を抽象化した少数派ゲームの計算機シミュレーションによる分析を行い、参加周期の異なるプレイヤが混在するケースで長期トレーダーが有利になる条件を導出した。[石川竜一郎]資産取引実験を行った。実験において、異なる論理能力を持つ被験者が同一市場で資産取引を行なった時、資産価格にどのように影響するか、資産価格バブルを助長するかという事が焦点になった。同一市場に参加する被験者の論理能力が多様であるほど、資産価格バブルは助長されるということが分かった。[竹内あい]帰納的ゲーム理論と献血に関する実験研究を行った。前者に関して、論文の修正を行っている。分析単位をグループから個人に切り替え、個人レベルでの行動の収束先に関する分析を行った。金子守・鈴木信行・Tai-Wei Hu (海外共同研究者)は認識論理学において加算無限個の命題の連言（＆）を許す体系を発展させている。既存の体系とは異なり、有限的体系から構成されており、さらに体系の大きさの順序により階層をなすことが示されている。これにより、「共通認識」の取り扱いが単純になり、帰納的ゲーム理論との結びつきも明らかにすることが期待できる。金子守、竹内あい（分担者）、船木由喜彦（分担者）、Jeffrey Kline (海外共同研究者)は帰納的ゲーム理論における実験に結果の分析に長い時間を要したが、研究の方向性が明確になってきており、今年度中に論文にまとめられる目途がついてきた。石川竜一郎（分担者）は帰納的ゲーム理論において、経験からの知識の改訂・社会的分業の問題を考察してきて、これと分析哲学との関連を考察してきた。これによって、帰納的ゲーム理論の分析哲学・社会経済学における立場が明確になりつつある。これも個人認識のあり方（社会構造の個人の学習）を明確にする。以上は当研究課題の参加者達の研究の一部である。それらを国際研究集会と国内の研究集会において報告・議論し、国際的な研究者達によって吟味してもらった。具体的には、平成28年10月7日～8日に、第9回太平洋ゲーム理論カンファレンスを早稲田大学SGU実証政治学拠点との共催で開催した。ニューヨーク州立大学ゲーム理論センターの Pradeep Dubey教授を招聘し、Peter Wakker教授（Erasmus大学）、Rabah Amier（Iowa大学）教授による報告など最先端の研究成果を発表し、議論を行った。平成29年3月21日に、研究会：社会慣習・行動パターン・限定合理性を開催し、社会経済問題と限定合理性に関して議論を行った。制度・認識・社会正義という研究対象は今まで通りとするが、本プロジェクトの過去3年間の進展に合わせて、より詳細にまたより広い視野から制度・認識・社会正義の研究を進めたい。特に、限定合理性は当プロジェクトにとって重要なキー概念になることが昨年度の研究で分かってきている。ただ、この場合、限定合理性はH. Simon の限定合理性より広い意味で理解する。例えば、論理能力、個別事象の記憶、パターンの記憶等である。これは社会制度や広い意味での社会的分業の問題に関連している。今まで、社会正義の問題は殆ど手を付けてこなかった。平成28年度中に進展させた確率使用を限定した期待効用理論がほぼ完成したので、これを基礎にしたナッシュ社会厚生関数の理論を発展させる予定である。これにより社会正義の問題を語る基礎付けが可能になる。具体的には、ナッシュ社会厚生関数が基礎とする効用の基数性を導くのに必要な確率は、どの程度必要かが分析できる。実は、あまり深い（詳細）確率は必要としないことが導かれると予想している。また、賃貸住宅市場論での市場均衡価格の分布の計算機プログラムで、計算を視覚的にするプログラムの開発を金子守・秋山英三が計画している。これは同時に市場均衡理論に出てくる経済主体を「限定合理性」の立場から論ずることにもなり、市場均衡理論と帰納的ゲーム理論との接続を可能にする。平成29年9月に第10回太平洋ゲーム理論カンファレンスを開催する予定である。3年間の成果を報告して、参加者からコメント・評価をもらい、今後の研究につなげたい。このコンファレンスの参加者は、国内・国外から20名程度である。その他、定期的にゲーム理論・経済学・認識論理学の研究会を計画している

The purpose of this project is to construct inductive learning theory to describe learning process of bounded rational decision makers, which is based on their experiences. We focused on experimental asset markets, and analyzed their behavioral data for the following purposes: (i) Classification of subjects’ rationality, (ii) Extraction of effects of their higher order beliefs, (iii) Their predictions of future prices, and (iv) Effects of their heterogeneous experiences. As a result, we found that the bubble of asset prices is facilitated by heterogeneous cognitive abilities, and is due to subjects’ higher order beliefs. Furthermore, we also observed that the experienced subjects in experimental markets adaptively learn price paths in their present markets. In addition, we show a new methodology of experimental assets markets to control subjects’ incentive when they have more incentives

In this study, I have both theoretical and experimental studies on mark-to-market accounting systems. In the theoretical work, I show that the system encourages financial contagion, and that banks will go bankrupt even when the banks prepare the liquidity crises.In the experimental work, we have experiments to study how subjects recognize the public information of accounting. The, we have the following results: (1) the subjects who have more information do not necessarily obtain the higher rewards, and (2) When the subjects obtains the higher rewards, they have the better strategy due to the information advantage

In order to take account of the impact of voters' spatial location, this project is an attempt to characterize the gap between voting location and optimal one by using a simple public facility location model. First, we derive the necessary and sufficient conditions that the voting outcome coincides with efficient one. Then we give simple examples which reveal the inefficiency of the voting solutions. Third we theoretically demonstrate that the network model can reduce to the linear model, indicating that our findings are applicable to network location problems. Finally, we apply our model to an actual city hall location problems in Kanto Region, Japan, using visualization techniques based on GIS

The productions of this project are classified into either theoretical ones and experimental economic ones. In theoretical productions, we propose a framework of dynamic logic for inductive game theory. In addition, we study how the difference of information decision makers obtain makes an impact on decision making and how the social welfare varies.In experimental economic productions, we study how subjects expect asset prices through their experiences and cognitive abilities. This is an application of the theoretical products for experimental asset markets. As a result, we found the subjects who get higher score in a cognitive ability test make a greater impact on strategic uncertainty

In this research project, we have studied social institutions from those in the broad sense such as customs, and moralities to those in the narrow sense such as legal and economic institutions having written codes. Studies of those enable us to make comparisons between societies with different cultural backgrounds. Taking those problems as concrete objects, we conducted research extensively on inductive game theory, epistemic logic, and experimental studies

As a theoretical model of competitive transaction network, a model of ‘Wicksellian Transaction Networks’ is constructed wherein various competitive transaction structures affect resulting outcome of competition within the network. A Wicksellian Transfer Game is introduced on the basis of a model of Wicksellian transaction networks and power distributions among buyers and sellers as their number in the game varies are analyzed

This project analyzes market trade under uncertainty by focusing "private information" and "higher order beliefs." I analyze how acquisition of private information gives impacts to traders' higher order beliefs and asset prices. In addition, I theoretically show how the difference of economic welfare by how a policy maker understands traders' payoffs

We focus not only on “people with disability (PwD)” in its narrow sense, but also on those stuck between different systems, including people with long standing health problems and people with facial disfigurement, and study social barriers in their way. In addition to theoretical analysis using game theory and disability studies, we conducted statistical surveys of PwD in cooperation with disability groups and local government, of firms, of patients with long standing health problems. We also conducted statistical survey in foreign countries such as Nepal and Philippines