Volume 185

Under the background of intensifying competition in the health insurance industry, it is of great significance to analyze the financial status and development potential of enterprises for investors to make decisions. This paper takes the four major American Health insurance companies (UnitedHealth Group, Centene Corporation, Elevance Health Corporation and Cigna Group) as the research object. The purpose of financial index analysis (profitability, growth ability, valuation level) is to evaluate the competitive potential of enterprises and provide basis for investment decisions. CI achieves high growth through refined cost control and business expansion, while UNH is slowing down due to diminishing marginal effects of scale, and other companies (such as CNC) have potential investment opportunities due to low valuations. Investors should prioritize Cigna Group (CI) because of its high profitability, strong growth and low valuation with long-term return potential, while keeping an eye on industry policy changes and market volatility risks. There is still some room for improvement in this study, which needs to be improved gradually in the follow-up work.

Show more

This study examines whether and how data asset disclosure affects ESG rating divergence. Based on a sample of A-share listed companies in China from 2007 to 2023, this study empirically examines the impact of data asset disclosure on ESG rating divergence using a fixed-effect model and multiple robustness tests. The results indicate that data asset disclosure significantly reduces ESG rating divergence. Mechanism tests indicate that data asset disclosure reduces ESG rating divergence through enhancing information transparency, improving green disclosure practices, and lowering earnings management. Heterogeneity tests find that the mitigation effect of data asset disclosure is more pronounced in non-state-owned enterprises and institutional firms with lower investor ownership. This study offers empirical evidence for firms aiming to reduce ESG rating divergence via strategic data disclosure.

Show more

Traditional single-risk assessment models perform inadequately with imbalanced data. This study aims to construct a high-precision and stable credit default prediction model to address both data imbalance and model generalization issues. This study aims to develop a high-precision credit default prediction model to address the challenges of imbalanced data and model generalization. By applying SMOTEENN hybrid sampling to optimize data distribution and integrating it with a Stacked Logistic Regression Ensemble (LR-Stacking), the model combines the predictive advantages of XGBoost, CatBoost, and Random Forest through a meta-learning layer. This approach effectively enhances the recall rate for imbalanced data and improves model generalization. Empirical results demonstrate notable improvements: the model achieves a recall rate of 0.72 for default samples, maintains an AUC score of 0.7341, balances risk coverage and prediction precision, and particularly enhances model stability. By integrating the Stacked Logistic Regression Ensemble Model (i.e., LR-Stacking), we use a meta-learning layer to combine the predictive strengths of XGBoost, CatBoost, and Random Forest. This approach ultimately enhances recall rate and model generalization performance in imbalanced data scenarios.

Show more

With the acceleration of urbanization and the diversification of residents’ travel needs, the modes of transportation have been constantly evolving. Traditional taxis, as an important part of urban public transport, have long provided convenient travel services for citizens. However, with the rise of ride-hailing services, this emerging mode of transportation has quickly gained a foothold in the market due to its high efficiency and convenience. Therefore, this paper analyzes the operational differences between traditional taxis and ride-hailing services in big cities, focusing on aspects such as pricing mechanisms, hailing methods, service coverage, and service quality. Traditional taxis have stable prices based on the meter, with extensive service coverage suitable for remote areas and special circumstances. Ride-hailing services use dynamic pricing with significant price fluctuations, and are more convenient for hailing, with better service in urban areas and during peak hours or specific time periods. Traditional taxi drivers follow certain service standards but have varying vehicle conditions, while ride-hailing drivers have mixed quality, though some platforms have strict assessments. These differences provide passengers with diverse choices and offer references for industry development, aiming to support the traffic management authorities in formulating policies, optimizing the transportation system, and enhancing passengers’ travel experiences.

Show more

With the increasingly fierce competition in the financial industry, customer churn prediction has become a key research topic. Accurate prediction of which customers are more likely to churn can help banks take timely retention measures to reduce business losses. This paper adopts a data-driven approach and uses the public bank customer churn dataset to deeply analyze the distribution of data characteristics and deal with the problem of data imbalance, and proposes a customer churn prediction method based on stacked ensemble model. In this study, random forest, XGBoost, CatBoost and LightGBM were used as the basic model, and XGBoost was used as the meta-learner to establish a two-layer stacked ensemble framework. Compared with the traditional single model and simple ensemble methods, the experimental results show that the proposed method is significantly ahead in Accuracy, Recall, AUC, F1-score and other indicators, which verifies its advanced and precise capabilities in customer churn prediction.

Show more

The paper aims to compare the comprehensive capabilities of four renowned pharmaceutical companies - AbbVie, Johnson & Johnson, Pfizer, and Merck & Co. through multi-dimensional analysis. It begins by outlining the current development status and trends of the global medical and pharmaceutical industry, thereby providing essential contextual support for the subsequent analysis. Subsequently, it offers detailed profiles of each company, encompassing their developmental history, core business areas and market positioning. Using an integrated approach whicht combines qualitative and quantitative methods, this paper assesses the performance of each company across various dimensions, including R&D capability, operating efficiency, marketing strategies, financial health and social responsibility. The findings indicate that Pfizer has remarkable proficiency in R&D and marketing. While Johnson & Johnson has advantages in diversification and brand influence. AbbVie established its leading position in the field of immunology. Merck & Co. achieved significant accomplishments in market expansion and cost control. Despite their distinctive strengths, each company encounters unique challenges worth paying attention to. The paper summarizes the comprehensive ranking of these companies and puts forward specific suggestions for their future development.

Show more

In this era of advanced payment methods, credit cards have become an indispensable financial instrument for individuals and business alike. The rapid development of credit card business has led to an escalation in credit card default issues, resulting in significant economic losses and risks for financial institutions. This study utilizes the Kaggle credit card default dataset to conduct credit card default prediction using the Random Forest model, with comparative analysis against the Logistic Regression model. The research findings demonstrate that the Random Forest model outperforms the Logistic Regression model across various evaluation metrics, including accuracy, recall, F1 score, ROC curve, and AUC value, particularly excelling in handling nonlinear relationships and high-dimensional data. Through feature selection, the study identifies key characteristics influencing credit card default, such as repayment status, credit limit, and bill amount. The research indicates that the Random Forest model can effectively identify potential default customers, assisting financial institutions in reducing default risks and enhancing risk management capabilities.

Show more

The insurance industry is undergoing a transformative evolution as artificial intelligence (AI) revolutionizes traditional actuarial practices. Although AI gets to enhance predictive models, operational efficiency and product innovation, however, its integration also brings major challenges, including ethical dilemmas, regulatory obstacles, algorithm opacity and workforce displacement. This article examines the practical challenges of applying artificial intelligence in the field of actuarial, especially on data governance, transparency and accountability. Through the industry case studies, it had analyzed the impact of artificial intelligence on underwriting, claim handling, fraud detection and compliance, meanwhile emphasis the emerging risks such as decision-making bias and regulatory non-compliance. This article research result shows that, the successful application of artificial intelligence requires a robust system of technical governance framework, an explainable models, employee skills retraining, and proactive supervision and coordination. Suggestions include establishing an ethical supervision mechanism, improving the transparency of algorithmic decision-making, and promoting continuous learning to make up for the skill gap. The present paper advocates for a harmonious blend - maximizing the advantages of artificial intelligence while keeping the disadvantages at bay - towards creating responsive and fair outcomes. In the end, the present paper makes pragmatic recommendations to insurance firms, actuaries, and policymakers to help insurance firms, actuaries, and policymakers cope with the convergence of the fast-evolving world of artificial intelligence and actuarial science and appeals for protection of the trust of the people and maintenance of professional ethics during this period of accelerated technology advancement.

Show more

With the rapid development of urban rail transportation, metro commercial streets have gradually become an important way to drive urban economic development. As seen in many movies, dramas, and promotional material, prosperous scenes presented by Tokyo’s metro commercial streets show a mature development model. In order to explore the effectiveness of this development model and its significance for other cities, this paper takes Tokyo and Hangzhou as case studies and analyzes the development status and differences of metro commercial streets in the two cities from the perspectives of urban planning, economic development, passenger flow, and development and operation models. It is found that Tokyo has effectively promoted the integration of commercial streets and urban space through the mature “transportation + business” model, forming a number of well-functioning urban commercial hubs. Although Hangzhou has been developing rapidly in recent years, it is still in an early stage due to several practical constraints. In addition, Hangzhou has the unique advantage of combining commercial areas with scenic attractions, which is rare among Chinese cities and should be further leveraged. Finally, this paper proposes that Hangzhou can learn from the Tokyo model to promote the comprehensive development of commercial areas around metro stations, create integrated spaces for shopping, working, and living, and foster a more dynamic interaction between rail transit and urban economy.

Show more

Against the backdrop of rapid advancements in fintech and the continuous expansion of credit markets, traditional credit risk assessment methods have revealed significant limitations. Machine learning methods offer new opportunities for credit risk management. This study focuses on applying machine learning methods to credit risk management. We utilize a credit risk dataset from Kaggle (Default of Credit Card Clients Dataset) to analyze and compare the performance of logistic regression, decision tree, and random forest models across multiple dimensions, including accuracy, recall rate, and interpretability. The results demonstrate that the decision tree model exhibits comprehensive performance in credit default prediction. Future research could incorporate diverse data types, develop visualization tools, establish real-time monitoring and dynamic updating systems, and extend applications across industries to enhance the accuracy and foresight of credit risk assessment, thereby promoting the widespread adoption of machine learning in financial risk management. The research holds theoretical significance and offers practical technical solutions for real-world lending operations.

Show more