专利转化特征精准识别与预测——以人工智能芯片为例

探讨前沿科技领域专利转化特征并对其进行精准识别与预测，对于我国破解“卡脖子”技术难题及实现科技自立自强具有重要意义。选取人工智能芯片专利领域，采用机器学习算法测度最优转化预测方案，分析全球范围内主要国家或地区专利成功转化影响因素，从企业/高校、国内/国际等不同层面总结专利成功转化的主要特征。结果发现：随机森林算法预测效果较好，人工智能芯片领域专利转化概率服从对数曲线分布，影响高校/企业、国内/国外专利转化特征的因素有所不同。最后，提出高校/科研机构应注重高价值专利维持和团队合作、企业应提升专利技术质量和撰写质量等政策建议。     

中国机构投资者的学习机制研究——理性贝叶斯还是简单强化式学习

理论文献通常假设机构投资者遵循理性贝叶斯法则更新其信念,在此学习模式下,盈亏经验本身并不能直接影响机构行为。但该假设的合理性尚未得到实证研究的充分支持。中国新股发行中的抽签分配制度为检验上述命题提供了比较理想的随机实验机会。本文基于新股抽签分配数据,系统检验了随机的盈亏经验对机构投资行为的影响,结果发现:(1)机构投资者显著受制于幼稚的强化式学习机制的影响,即通过随机抽签在前期获得新股分配的机构(处置组)相对未获配机构(控制组),其下期参与新股申购的概率显著提高,并且前期收益率能够有效地强化这种盈利经历与参与概率之间的正向关系。(2)盈亏经验能够显著改变专业机构的估值信念,即前期通过随机抽签获得高收益体验的机构相对未获配机构,在后续新股询价过程中给出了显著更高的报价上调水平。(3)盈亏经验对机构行为产生影响的一种可能渠道是借助机构投资管理人的强化式学习过程,基于基金经理个人特征变量的调节机制研究表明,丰富的长期从业经验、高学历的教育水平以及多位基金经理的相互竞争都能在一定程度上缓解盈亏经历对基金行为的影响。本文基于随机实验的设计为经验与行为之间的因果关系提供了可信证据,证实了即使是被奉为理性投资者代表的专业机构也会受制于简单强化式学习的显著影响。     

Explainable Artificial Intelligence (XAI) in auditing

Artificial Intelligence (AI) and Machine Learning (ML) are gaining increasing attention regarding their potential applications in auditing. One major challenge of their adoption in auditing is the lack of explainability of their results. As AI/ML matures, so do techniques that can enhance the interpretability of AI, a.k.a., Explainable Artificial Intelligence (XAI). This paper introduces XAI techniques to auditing practitioners and researchers. We discuss how different XAI techniques can be used to meet the requirements of audit documentation and audit evidence standards. Furthermore, we demonstrate popular XAI techniques, especially Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive exPlanations (SHAP), using an auditing task of assessing the risk of material misstatement. This paper contributes to accounting information systems research and practice by introducing XAI techniques to enhance the transparency and interpretability of AI applications applied to auditing tasks.     

Forecasting solar stock prices using tree-based machine learning classification: How important are silver prices?

Solar energy is one of the fastest growing sources of electricity generation. Forecasting solar stock prices is important for investors and venture capitalists interested in the renewable energy sector. This paper uses tree-based machine learning methods to forecast the direction of solar stock prices. The feature set used in prediction includes a selection of well-known technical indicators, silver prices, silver price volatility, and oil price volatility. The solar stock price direction prediction accuracy of random forests, bagging, support vector machines, and extremely randomized trees is much higher than that of logit. For a forecast horizon of between 8 and 20 days, random forests, bagging, support vector machines, and extremely randomized trees achieve a prediction accuracy greater than 85%. Although not as prominent as technical indicators like MA200, WAD, and MA20, oil price volatility and silver price volatility are also important predictors. An investment portfolio trading strategy based on trading signals generated from the extremely randomized trees stock price direction prediction outperforms a simple buy and hold strategy. These results demonstrate the accuracy of using tree-based machine learning methods to forecast the direction of solar stock prices and adds to the broader literature on using machine learning techniques to forecast stock prices.     

Gender differences in the first course in accounting: An achievement goal approach

We examine whether gendered patterns can be observed in first-year students' achievement goals in an introductory accounting course; a question largely overlooked by prior literature. This investigation is motivated by perceptions of accounting as a masculine field involving gender role stereotypes and business schools as competitive and performance-oriented environments. Our findings suggest that male students tend to adopt performance-approach goal, implying that they are more competitive than female students, and that their performance is thus driven by a desire to outperform others. Our findings further suggest that male students' expectations of learning accounting are higher than those of female students. The expectations explain the gender differences in the performance-approach goal. Finally, we find that this performance-approach goal mediates gender differences in course performance depending on the mode of assessment; male students received higher grades for exams but not for teamwork. Overall, our study highlights the importance of considering contextual aspects related to competitiveness, masculinity, and the mode of assessment on an accounting course when addressing students’ achievement goals and expectations of learning accounting. We thus contribute to the understanding of how learning environment, accounting pedagogy, and the broader field of professional accounting intersects with individual student attributes, creating differential learning outcomes.     

Economists in the 2008 financial crisis: Slow to see,fast to act

We study the economics- and finance-scholars’ reaction to the 2008 financial crisis using machine learning language analyses methods of Latent Dirichlet Allocation and dynamic topic modelling algorithms, to analyze the texts of 14,270 NBER working papers covering the 1999–2016 period. We find that academic scholars as a group were insufficiently engaged in crises’ studies before 2008. As the crisis unraveled, however, they switched their focus to studying the crisis, its causes, and consequences. Thus, the scholars were “slow-to-see,” but they were “fast-to-act.” Their initial response to the ongoing Covid-19 crisis is consistent with these conclusions.     

机器学习在人力资源管理领域中的应用研究

本研究关注的是机器学习在人力资源管理领域的应用,通过目前已有研究的梳理,以了解在人力资源管理领域中人工智能/机器学习的应用程度和研究热点。同时,基于管理实践,本研究提出了人力资源管理者如何利用算法,以有价值的方式处理和解释数据,将其真正应用于人力资源管理的六大模块工作中,以降低管理成本,提供不断增长的优势和潜力。最后,在现有学术研究和管理实践的基础上,本研究列举了机器学习在人力资源管理领域应用可能面临的挑战,以及未来的研究方向,以期为后续的研究提供一定的参考。     

Does air pollution influence music sentiment? Measuring music sentiment by machine learning

Air pollution has imposed significant negative effects on individuals’ well-being, including citizens’ sentiment levels. To test this claim, we investigate the impact of air pollution on Chinese urbanites’ music sentiments. The analysis is based on a unique dataset of high-frequency music consumption records from a music platform in China from October 13th, 2019 to January 7th, 2020. Using machine learning algorithms, songs on this platform are divided into cheerful songs, melancholy songs and other categories, by which a music sentiment index (MSI) is generated at city-daily level. By matching MSI and daily air quality, this study finds that the MSI declines during highly polluted days, indicating that: on highly polluted days, citizens tend to enjoy melancholy songs over cheerful ones. In addition, this effect becomes more remarkable when the Air Quality Index (AQI) score is above 200, a critical point for “heavily polluted” and “severely polluted” days.     

Social media networks,fake news,and polarization

We study how the structure of social media networks and the presence of fake news affects the degree of misinformation and polarization in a society. For that, we analyze a dynamic model of opinion exchange in which individuals have imperfect information about the true state of the world and exhibit bounded rationality. Key to the analysis is the presence of internet bots: agents in the network that spread fake news (e.g., a constant flow of biased information). We characterize how agents’ opinions evolve over time and evaluate the determinants of long-run misinformation and polarization in the network. To that end, we construct a synthetic network calibrated to Twitter and simulate the information exchange process over a long horizon to quantify the bots’ ability to spread fake news. A key insight is that significant misinformation and polarization arise in networks in which only 15% of agents believe fake news to be true, indicating that network externality effects are quantitatively important. Higher bot centrality typically increases polarization and lowers misinformation. When one bot is more influential than the other (asymmetric centrality), polarization is reduced but misinformation grows, as opinions become closer the more influential bot’s preferred point. Finally, we show that threshold rules tend to reduce polarization and misinformation. This is because, as long as agents also have access to unbiased sources of information, threshold rules actually limit the influence of bots.     

Forecasting cryptocurrency returns with machine learning

This article employs machine learning models to predict returns for 3703 cryptocurrencies for the 2013 – 2021 period. Based on daily data, we build an equal (capital)-weighted portfolio that generates 7.1 % (2.4 %) daily return with a 1.95 (0.27) Sharpe ratio. We obtain an out-of-sample R² of 4.855 %. Our results suggest that cryptocurrencies behave like conventional assets than fiat currencies since variables, including lagged returns, can predict future returns. As assets, cryptocurrencies are not weakly efficient, and production costs do not determine their prices. Returns for small cryptocurrencies are more predictable than larger ones. The predictive power of the 1-day lagged return is stronger than all other features (predictors) combined. The results offer new insights for crypto investors, traders, and financial analysts.