Published papers

Japan is suffering from a serious labor shortage due to the economic expansion anddemographic change in the non-regular employment of the labor market. In particular, non-regularemployees account for 80% of employees in the food-service industry is a critical issue. The purpose ofthe present study is to analyze a mechanism of job matching focusing on non-regular employmentthrough agent-based modeling.

近年、情報に対する関心が高まっている。個人が情報を獲得・処理するには様々な技術が必要とされる。このような中、近年、技術を有する主体と、有せず情報処理ができない主体との格差が広がる可能性がある。本研究では、取引市場を対象とし、主体が保有している情報の格差が、個々の主体及び市場全体に及ぼす影響についてエージェント・ベース・モデルを通じ分析する。本研究では、情報の保有が特に有益となる可能性のある相対取引に焦点を当て分析を試みる。具体的には、sugarscapeのモデルを用い、個人（エージェント）の情報格差を移動行動と取引条件に反映する。今後の課題として、金融市場への応用とよりエージェント学習の高度化を検討する。

While the importance of financial education increases in recent years, the technique for deepening an understanding of finance theory is needed. In this research, we analyize learning method of the finance theory about the investment project selection and capital structure determination using the business game technique. As a result of analysis, the participant understood the investment project selection method and interesting phenomena - an understanding progresses about the method of determining the capital structure which raises capital stock value - were seen. These results show the effectiveness of the business game technique to study of finance theory.

In a policy simulation for the life planning of elderly, the possibility of future asset depletion is analyzed based on clusters generated from specific attributes and segments. However, a room for improvement exists in the detailed understanding of the attributes that branch the simulation re-sults and comparative examination of possible metrics. In this study, we propose a life planning support system that combines the existing method with social simulation log analysis. This system has the following steps: I) Performs simulations based on clusters generated from the feature analy-sis of individual data. II) Grasps the overall pattern of possible results by comprehensive simula-tions based on the individual data and hierarchical classification of the obtained simulation logs. III) Considers the possible measures by comparing the results of I) and II). The main results of this study are as follows: a) Ability to grasp the threshold of the attribute that separates asset depletion and non-depletion. b) Ability to specify countermeasures that can be taken for each cluster along with numerical goals comprehensively and semiautomatically. The proposed system enables de-tailed advice from financial institutions by providing knowledge directly linked to life planning, which leads to an improvement in the effectiveness of industrial applications.

This paper proposes a method of building a sentiment dictionary using only news and stock price data for textual analysis in finance. To obtain word polarity from stock price fluctuations, we calculate stock price returns following announcements of news articles. We constructed learners with support vector regression, using stock price returns as supervised labels of news articles, and built a sentiment dictionary by extracting word polarity from learners. Furthermore, we examined whether our sentiment dictionary is effective in classifying news articles as negative or positive. We found that our sentiment dictionary is also effective in classifying news articles provided by other news media other than news media we employed in constructing the algorithm. In addition, we found that it is difficult to classify news articles on a date that is two trading days away from the news announcement date.

While general investment behaviors on the news are comprehended through market movements, individual investor's actions are rarely observed. This research focused on stock trading on the October 31, 2014 and empirically analyzed investor behavior against the released news on that day. Records that we analyzed reflected High-Frequency Trading (HFT), so that we can target short-term investor behavior. As a result, divergence between principal and intermediary (agent) investment were determined as well as individual investment behavior.

組織研究では，ビジネスケースや組織シミュレーションを用いるアプローチがある．自然言語により記述されたビジネスケースや多数のシミュレーション・ログには，解釈や認識の多義性が生じうる．そのため，ケースやモデルの作成者や分析者の間で，当該多義性を低減することが重要となる．なぜならば，双方の接近法の間での議論やコミュニケーションを活発化し，効率的な事例比較を行うために不可欠だからである．本稿では，ビジネスケースに係る抽象的記述言語を用いることで，実在のビジネスケースと組織シミュレーションのシミュレーション・ログにつき，それぞれの意思決定プロセスを可視化し，形式的に記述・表現しうることを例示する． In organization research, there are approaches using business cases and organizational simulations. Business cases described in natural language and numerous simulation logs may have ambiguity in interpretation and recognition. Therefore, it is important for creators and analysts of cases and models to reduce such ambiguity. This is because activating discussions and communication between the two approaches, which is indispensable for efficient case comparisons. In this paper, we demonstrate that each decision making process are able to be visualized and formally described for actual business cases and simulation logs of organizational simulation by using an abstract description language for business cases.

近年,金融教育の重要性が高まる中,ファイナンス理論の理解を深めるための手法が必要とされている.本研究では,ビジネスゲーム手法を利用した投資プロジェクト選択と資本構成決定に関するファイナンス理論の学習を目的とした分析を実施した.分析の結果,参加者は投資プロジェクト選択法を理解し,株主資本価値を高める資本構成の決定法について理解が進むなど,興味深い現象が見られた.これらの結果は,ファイナンス理論の学習へのビジネスゲーム手法の有効性を示すものである.

As a case study of around the Shibuya station, the multiple regression analysis of housing, office and commercial rent was examined with building and location measures including visibility and integration value by space syntax theory as candidate factor variables. After the factor variables were adopted by t-test, we compared and considered the multiple regression models of housing, office and commercial use, using the factor ranking by t-value. The result showed integration value of space syntax measures were adopted in all models, although the ranking was different in each models.

This paper proposes a unique estimation method of Jump Factor with high frequency data. Our method employs bi-power variation technique and principle component analysis. Proposed method is unique and exhibits a new measure for financial risk. In this study, we calculated four Jump Factors in United States and Germany and integrated them through principle component analysis. In addition, we could confirm the influence of Jump Factors on financial markets by using the proposed method.

Since systemic risk in recent financial crisis has received attention, a wide variety of studies on how financial regulations should be reformed to control such risk have been progressed. Among these, mathematical and computational studies have analyzed the way how the borrowing and lending banks and the borrowers go bankrupt in the chain via interbanking network. This study deals with a chain of bankruptcies of financial institutions endogenously caused by deterioration of their financial situations from the changes in prices of risky assets by focusing on the macro-level collapse caused by the shock stemming from the general market risk factors addressed in theoretical studies after the crisis in 2008. For this purpose, the authors develop an agent-based simulation platform and then examine how current systemic management regulations affect corresponding bankruptcies. The main findings are as follows: First, pertinent management regulations are dependent on the market environment. Second, some combination of management regulations may increase the possibilities of bankruptcies due to more sensitivity toward the market change.

News articles distributed in financial markets are essential information that affects asset valuation. Many studies have analyzed the impact of delivered news articles on stock price fluctuations. However, the number of news articles is limited, and the limit on the number of data that can be obtained affects the accuracy of analysis using deep learning. In this study, we constructed a news evaluation model based on stock price fluctuation before and after news article distribution and tried to improve the accuracy of the model by generating news articles through GPT-2. The analysis target was a Japanese electronics manufacturer, and analysis was performed using a news evaluation model. As a result of the analysis, we found that there is a possibility that the generation of news articles through GPT-2 can improve the accuracy of the news evaluation model.