This paper uses a 2016–2024 individual-level panel to test whether COVID-19 changed public attitudes toward women’s participation in the workplace. Using fixed-effects and fixed-effects ordered-logit models, the authors find that (1) attitudes became more positive after COVID-19, (2) older cohorts shifted from more negative to more positive views post-pandemic, and (3) married men and certain worker-status groups exhibited distinct changes; these results are robust across specifications [page::1][page::11][page::13]
This paper proposes a benchmark framework that generates 9,500 mathematically well-defined portfolio-optimization multiple-choice problems to evaluate LLMs' quantitative decision-making in asset allocation, comparing GPT-4, Gemini 1.5 Pro, and Llama 3.1-70B across objectives (variance, return, Sharpe, MDD, CVaR), constraint types, and distractor-generation methods, and finds model-specific strengths (GPT: risk objectives; Gemini: return-focused) and shared limitations on multi-criteria optimization such as Sharpe and CVaR [page::0][page::5].
This paper develops an expected-utility framework to quantify the cost of time (COT) and the cost of time variability (COTV), and derives tight upper bounds on COTV relative to COT under different preference and process assumptions; notable results include COTV/COT ≤ 1/2 for quadratic-utility users under a Poisson (exponential) service process and the general expression showing dependence on CV, relative risk aversion (RRA) and relative prudence (RP) [page::0][page::12][page::14]
本研究通过一项 2×3 线上随机实验,考察“已知风险(risk)”与“模糊不确定性(ambiguity)”下个人数据泄露威胁对用户接受 AI 个性化推荐的影响,并将经济性成本与非货币隐私伤害分离测量;结果发现:在已知风险情境下,隐私威胁并不降低 AI 采纳率,但在模糊概率情境下,隐私威胁显著降低个性化采纳(尤其对敏感人口统计数据方向性更强),且被试普遍愿付费购买能完全消除泄露不确定性的隐私标签(存在过度出价现象)[page::1][page::5][page::6][page::8]