In this paper, we explore a novel model for pricing Chinese convertible bonds that seamlessly integrates machine learning techniques with traditional models. The least squares Monte Carlo (LSM) method is effective in handling multiple state variables and complex path dependencies through simple regression analysis. In our approach, we incorporate machine learning techniques, specifically support vector regression (SVR) and random forest (RF). By employing Bayesian optimization to fine-tune the random forest, we achieve improved predictive performance. This integration is designed to enhance the precision and predictive capabilities of convertible bond pricing. Through the use of simulated data and real data from the Chinese convertible bond market, the results demonstrate the superiority of our proposed model over the classic LSM, confirming its effectiveness. The development of a pricing model incorporating machine learning techniques proves particularly effective in addressing the complex pricing system of Chinese convertible bonds. Our study contributes to the body of knowledge on convertible bond pricing and further deepens the application of machine learning in the field in an integrated and supportive manner.

This study explores the following two aspects: (i) the impact of Environmental, Social, and Governance (ESG) scores and corporate ownership characteristics on the performance of Chinese listed companies, and (ii) whether different ownership characteristics (state-owned, private, foreign) moderate the relationship between ESG participation and corporate performance. By analyzing a comprehensive sample of 4649 listed companies in China, we provide robust evidence that ESG participation and its three pillars (i.e., Environmental, Social, and Governance) can significantly enhance corporate performance, as measured by the accounting-based proxy return on assets (ROA). Moreover, our research findings reveal an important and novel discovery: in the Chinese market, ownership types have significantly different moderating effects on the relationship between ESG and corporate performance. Specifically, compared to state-owned enterprises and private corporations, foreign ownership exhibits a stronger moderating effect in enhancing the positive impact of ESG on ROA, followed by private corporations, while the moderating effect of state-owned enterprises is the weakest. This result provides new perspectives and empirical support on how ESG and ownership structure jointly affect corporate performance, offering references for future related research and policy formulation.

This study proposes the Double-Ensemble Learning Classification with SMOTE (DELC-SMOTE), a novel hierarchical framework designed to address the critical challenge of severe class imbalance in financial bond default prediction. The model integrates the Synthetic Minority Over-sampling Technique (SMOTE) into a two-phase ensemble architecture. The first phase employs introspective stacking, where six heterogeneous base learners are individually enhanced through algorithm-specific balancing and meta-learning. The second phase fuses these optimized experts via performance-weighted voting. Empirical analysis utilizes a comprehensive dataset of 10,440 Chinese corporate bonds (522 defaults, ~5% default rate) sourced from Wind and CSMAR databases. Given the high cost of both false negatives and false positives in risk assessment, the Geometric Mean (G-mean) and Specificity are employed as primary evaluation metrics. Results demonstrate that the proposed DELC-SMOTE model significantly outperforms individual base classifiers and benchmark ensemble variants, achieving a G-mean of 0.9152 and a Specificity of 0.8715 under the primary experimental setting. The model exhibits robust performance across varying imbalance ratios (2%, 10%, 20%) and strong resilience against data noise, perturbations, and outliers. These findings indicate that the synergistic integration of data-level resampling within a diversified, two-tiered ensemble structure effectively mitigates class imbalance bias and enhances predictive reliability. The framework offers a robust and generalizable tool for actionable default risk assessment in imbalanced financial datasets.

The continual reassessment method (CRM) has been an essential Bayesian finding design in phase I clinical trials. It utilizes all the information in observed data which contributes to its essential operational characteristics. However, the CRM has been criticized for its aggressive dose escalation. Model-assisted methods including BOIN, Keyboard, and mTPI improved the safety while retaining relative efficiency. In this paper, we propose four models combining the structure of the CRM and model-assisted methods. We show that these models could operate with comparable CRM performance through simulations. The results suggest that two of the proposed methods outperformed the traditional methods with a higher percentage of correct selection of true maximum tolerated dose. In addition, the interval-based approaches offered by the new models with greater flexibility regarding target toxicity achieved an improvement in the adaptability of the dose-finding process in clinical trials.

Background: It has been suggested that prenatal sex hormones may have a substantial impact on brain development and mental illness. The second-to-fourth digit (2D:4D) ratio, a critical retrospective marker for the secretion of prenatal sex hormones, can be measured and compared between healthy people and patients with schizophrenia as a direct way to support the hypothesis. Aim: This study aimed to examine whether sexual dimorphism is held in a 2D:4D ratio and further test the hypothesis that patients with schizophrenia have distinct 2D:4D ratios from healthy people. Setting: The study was conducted at a local hospital in a typical mid-size city in the eastern part of China. Methods: The study was performed by analyzing the 2D:4D ratios in both hands from the control group and patient group, respectively, with the application of statistical measurements (means, variances, and correlations), the independent-sample t-test, and the paired-samples t-test. In terms of the sample size, a total of 108 healthy participants were included in our study, comprising 33 men and 75 women, with a mean age of 30.55 years. Besides, there were 126 individuals with schizophrenia in the study, comprising 73 men and 53 women, with an average age of 44.2 years. Results: The results showed that the 2D:4D ratio of women in the control group (mean value of the left hand: 0.9650; mean value of the right hand: 0.9562) was significantly higher than that of men (mean value of the left hand: 0.9418; mean value of the right hand: 0.9298), indicating that the sexual dimorphism is held in the 2D:4D ratio (left hand p-value: 0.003<0.05; right hand p-value: 0.004<0.05). This effect, however, was not observed in the patient group. Instead, the female patients displayed more masculine digit ratios, whereas the male patients' hands displayed fewer masculine patterns. Conclusion: The results provide evidence that the pathogenesis of schizophrenia is related to abnormal prenatal sex hormone secretion. The research on the 2D:4D ratio as a biomarker should be able to help contribute to the literature on the pathogenesis of schizophrenia.

In the insurance industry, the actuarial team experiences significant challenges for pricing a competitive but also fair premium and keeping an accurate level of reserve, which leads inevitably to numerous adjustments over time and potentially several millions of US dollars in annual losses. Because the model and parameter uncertainties play key roles for actuaries, decision makers, and policymakers, the implementation of advanced mathematical and statistical techniques is highly required. Over the last two decades, applications of regime switching models to finance and economics have received strong attention among researchers and particularly among market practitioners. This paper attempts to consider how a linear arbitrary regime switching system in discrete-time framework could be applied to calculate the medium- and long- term reserves and the relevant premiums (abbreviated here as the P-R process) from the point of view of an insurer. In this direction, some recently developed techniques from linear robust control theory are applied to explore the stability, stabilization, and robust $H_{\infty }$ -control of a P-R system and the potential effects of abrupt structural changes in the economic fundamentals, as well as the insurer’s strategy over a finite time period. Sufficient linear matrix inequality conditions are derived for this treatment. Finally, a numerical example is illustrated.