Published papers

It is important to utilize intellectual capital to overcome the severe business environment. However, it is difficult to grasp the causal relationship between how to utilize them and how to improve business performance. In this study, we have developed a system that can elucidate the strategies of excellent companies through the analysis of their behavior in a complex business environment using an agent-based model. The system consists of three layers of business environment: external layer, decision-making layer, and choice of action layer. In addition, there are three types of investment targets: human capital, structural capital, and relational capital. Through this system, we can optimize investment strategies according to market conditions and recommend strategies that consider the behavior of competitors, which is the strength of agent-based models. In this paper, the construction of the system and a case study of analysis are presented. One of the novelties of this research is that we have developed a structure that allows us to consider the utilization of intangible assets from a quantitative perspective.

In this study, we attempt to extract corporate trends from news data. Specifically, in this analysis, we employed BERTopic to group news containing various topics into several categories. BERTopic is a tool for document clustering that utilizes BERT. The analysis focused on examining the relevance between topics related to products and corporate activities. The study targeted news headlines of specific chemical company listed on the Japanese stock market, covering a sample period of 24 years from 1996 to 2019. The results of the analysis revealed (1) the ability of BERTopic to classify target news into multiple groups and (2) the potential to efficiently obtain information about future corporate activities from the content of the news. These findings indicate the possibility of acquiring timely information about corporate trends through the analysis of news headlines using BERTopic.

This study attempts to identify the determinants of changes in real estate investment trust (REIT) unit prices at the time of seasoned equity offering (SEO) in the Japanese financial market. As Japanese REIT (J-REIT) announces SEOs, dividends, and borrowings at the same time, this analysis can consider each factor simultaneously. This study uses the event study method to conduct an ordinary least squares multivariate regression analysis (regression analysis) of the average cumulative abnormal return for 11 days during SEOs in J-REIT. We add the change in a borrowing ratio before and after an SEO and the dilution ratio of investment units from the perspective of the financing method as new candidate factors not used in previous studies. The sample period is from 2013 to 2017, and the number of the analyzed SEOs is 150. This study simultaneously analyzes the impact of the changes in dividends, the changes in borrowing ratios, and the dilution ratios on J-REIT unit prices at the time of SEOs and clarifies each impact. This study showed that investors were aware of the importance of the dilution impact from the perspective of net asset value as well as the dividend as profitability. Unlike previous studies, we also showed that the increase in the borrowing ratio negatively affected the unit price.

In this study, we analyzed the relationship between the technological distance calculated from the patent text data as an explanatory variable and the number of occurrences of outliers in new patents as an explained variable. This study aims to explore the technological distance that brings about innovation. We vectorized the patent text data using natural language processing called SCDV and calculated the technological distance of companies. PCA and Local Outlier Factor (LOF) were applied to calculate outliers. As a result of comparing the groups divided according to technological distance, we confirmed that the group with the shortest technological distance had a higher mean number of outliers than the other groups. Furthermore, similar results were obtained when the same comparison was made after classifying by company size. These results suggest that M&As between companies with similar technologies produce more innovation as outliers.

This study investigates the impact of heterogeneous time sensitivity and word-of-mouth effects on the economic efficiency of UAV-vehicle hybrid delivery systems. Our analysis reveals that an overemphasis on time-sensitive customers in route planning can lead to partial optimization, thereby reducing overall economic efficiency. This highlights the necessity of balancing time sensitivity with broader efficiency objectives. Future research should focus on developing algorithms suited for multi-vehicle and multi-UAV systems that incorporate heterogeneous customer characteristics.

This study quantitatively evaluates how improving delivery time prediction accuracy constructed using real-world traffic information impacts economic efficiency through computational simulation. It proposes a practical framework to optimize delivery operations and support data-driven decision-making. The results show that improving delivery time prediction accuracy led to a 2% increase in revenue and a 21% improvement in customer satisfaction, highlighting the economic and operational benefits of integrating real-world traffic information. This indicates that enhancing delivery time prediction accuracy can lead to economic efficiency gains and improved customer experience, enabling businesses to optimize delivery operations and improve competitive advantage. This research strengthens the foundation of data-driven management and provides a practical framework for establishing a competitive advantage.

Following the COVID-19 outbreak, unlike the plunge in traditional mainstream assets, crypto assets haveperformed very well. Indices that reflect changes in the crypto assets market have also grown in recent years. However,in terms of cryptocurrency index-related studies, there are not many, and the period analyzed is mostly three years. Inthis study, we analyze the impact of crypto assets on portfolio construction and attempt to clarify the risk-returncharacteristics of portfolios that include crypto assets as investment targets. We analyze historical data from July 2014to April 2020 based on the cryptocurrency index CRIX and six other traditional mainstream assets. By using theDCC-GARCH model, this study finds out the low dynamic correlation between the crypto assets and traditional ones.By using the mean-variance model, Cornish-fisher expansion, and T-copula CVaR approach to check the frontier lineand portfolio performance, we find out that crypto assets have the potential to improve the risk-return characteristics oftraditional portfolios.

To investigate the risks of financial markets is one of the critical issues in risk management This paper proposes an Agent-Based Model to clarify microscopic and macroscopic links between investor behaviors and price fluctuations in a financial market. The analysis presented in the paper focuses on the role that investors' overconfidence plays in the financial market. From the simulation study of the agent-based virtual market, we have found that (1) overconfident investors emerge in a bottom-up fashion in the market, and (2) these overconfident investors have the ability to contribute to the market in which the trading prices are coincide with theoretical fundamental values.

Using large-scale patent data, we analyze the impact of mergers and acquisitions (M&A) on innovation activities in companies. The analysis classifies M&A according to the trend of innovation output through their index clustering. In addition, we measure the technological distances among companies using patent document data, then apply the distances to the evaluation of M&A. As a result, we find two types of clusters: 1) one with the tendency to increase the postmerger innovation outputs; 2) the other with a trend to decrease innovation output around M&A deals. The former cluster tends to have a smaller innovation output scale around deals than the other clusters. Furthermore, compared to the other groups, the group whose postmerger innovation output decreases has the largest increase in stock price return after the deal announcement. This study’s major findings indicate that the combination of unstructured text data and machine learning methods applies to M&A and innovation.

Amid escalating global warming and the finite nature of energy resources, reevaluating fossil fuel usage while reducing greenhouse gas emissions is imperative. This study introduces and assesses a novel gas energy model integrating Ene-Farm household fuel cells, the J-Credit Scheme for monetizing emission reductions, and a Community-Based Rate Plan offering rate discounts based on collective CO₂ emissions. Utilizing an agent-based model, we simulate individual behaviors and interactions to evaluate community-wide energy consumption and CO₂ mitigation. Results indicate that combining Ene-Farm and J-Credit Scheme effectively complements Community-Based Rate Plan, fostering a mutually beneficial framework for gas companies, consumers, and the environment. Additionally, peer influence enhances emission reductions even in households without Ene-Farm installations.

The impact of mergers and acquisitions on corporate research and development has attracted considerable attention. We examine whether acquisitions of firms with AI patents mitigate post-merger declines in patent quality using computational text analytics and event-time benchmarking. We find that the 90th percentile of forward citations does not fall, that the average abstract length, our disclosure proxy, rises slightly, and that changes in lexical diversity are small, favoring mitigation over degradation. These findings suggest that artificial intelligence is likely to have a positive impact on the efficiency of research and development in the context of mergers and acquisitions.

Financial inclusion requires equitable access to financial services, yet linguistic barriers hinder non-native Japanese speakers. This study examines using large language models (LLMs) to generate Easy Japanese for financial literacy education and proposes a reference-free evaluation framework. The study assesses the effectiveness of LLMs through manual, automated, and machine-based evaluations using Zero-shot and One-shot prompting. The findings demonstrate that LLM-generated Easy Japanese enhances accessibility while preserving accuracy. Additionally, SHAP analysis of linguistic features across Japanese, Chinese, and Korean speakers reveals that hiragana, kanji, and particle usage influence text difficulty. This study provides a scalable solution for financial education and contributes to advancing financial inclusion.

News articles have significant impacts on asset prices in financial markets. A great number of attempts have been conducted to ascertain how news articles influence stock prices. News articles have been reported to contain sentimental and fundamental information that affects stock price fluctuations, and many studies have been conducted to evaluate stock price fluctuations using them as analytical data. However, the limitations in the number of available datasets usually become the hurdle for the model accuracy. This study aims to improve the analytical model’s accuracy by generating news articles using language generation technology. We tested whether the model that used the generated data was better than the trained model with real-world data. The model constructed in this research is a model that evaluates news articles distributed to financial markets based on the price fluctuation rate of stock prices and predicts and evaluates stock price fluctuations. This study labeled based on high-frequency trading data and generated news articles using a large-scale language generation model (GPT-2). Also, we analyzed and verified the effect. In this study, we succeeded in generating news articles using the large-scale language generation model and improving the classification accuracy. Our method proposed in this paper has great potential to improve text analysis accuracy in various areas.

This research analysed the impact of the disparity in forecast accuracy among investors in the financial markets. This analysis has found the following interesting phenomena: 1 depending on market environments, investors with good forecast accuracy do not always survive in the market 2 where the performance measurement period is short, there may be a negative influence such as deviation from the fundamental value of trading prices. These results suggest the need for examining various factors when considering the impact of information technologies such as the text mining technique on financial markets, and are of great significance from both practical and academic viewpoints. Copyright © 2010 Inderscience Enterprises Ltd.

Recently financial education has become more important because financial technology is highly developing and financial market is growing in importance. In this research, we apply Business Game method to financial education. Especially we focused on learning of asset allocation. As a result of intensive experiments, we found that (1) players learned not to take excessive risk through business gaming and (2) they recognize the importance of risk control by our experiments. These findings indicate that our approach is valid for financial education.

Bitcoin is expected to play a vital role in the web3 and is attracting attention. Still, it is notorious as a hotbed for various fraudulent transactions, which are on the rise. One of the characteristics of Bitcoin is its anonymity. Bitcoin allows tracing transactions, but the owner cannot be identified. In addition, bitcoin is considered more difficult to trace criminals than bank transactions because of the unlimited number of addresses that can be created. This study shows that it is possible to detect criminal transactions with known transaction patterns using supervised learning. Still, it is difficult to detect criminal transactions with unknown transaction patterns using supervised learning. We also show that even with anomaly detection methods, illicit transactions are difficult to detect. From the above, we believe that Bitcoin is vulnerable to illicit transactions.

In this paper, we use Agent-Based Approach to analyze how asset prices are affected by investors and investment systems that are based on Behavioral Finance. We build a virtual financial market that contains two types of investors: fundamentalists and non-fundamentalists. As a result of intensive experiments in the market, we find that (1) the traded price agrees with the fundamental value and the fundamentalists survive according to the principle of natural selection in the case that the market contains the same number of fundamentalists and trend predictors (investors who make trend prediction), (2) the traded price largely deviates from the fundamental value and the non-fundamentalists frequently obtain excess returns and therefore the fundamentalists are eliminated according to the principle of natural selection in the case that the proportion of trend predictors is extremely high or in the case that the investment ratio of the risk asset is restricted, and (3) the traded price largely deviates from the fundamental value in the case that the non-fundamentalists estimate the losses excessively, as pointed in Prospect Theory. These results indicate that the non-fundamentalists affect the traded prices and obtain excess returns also in real markets.

This study analyses the efficacy of a passive investment strategy - which is one of the most popular investment strategies in the asset management business - thorough agent-based modelling. As a result of intensive experimentation, the following conclusions were confirmed: (1) overconfident investors could achieve a positive excess return in the market where there are no passive investors. However, (2) even if overconfident investors have better predictive power, they couldn’t survive in a market where passive investors exist. These results suggest the effectiveness of a passive investment strategy. The results are of both academic interest and practical use.