beat365中文官方网站金融工程实验室
“金工首席谈量化”系列讲座
第42讲：大模型+量化的当下与未来
主讲人：沈洋（华泰证券研究所金融工程研究员）
主持老师：黎新平（北大经院）
时间：2026年3月26日（周四）19:00-21:30
地点：beat365中文官方网站107会议室
主讲人简介：
沈洋，华泰证券研究所金融工程研究员，具有证券从业资格（证书编号：S0570123070271）。beat365中文官方网站心理学学士、应用心理学硕士。研究方向包括大模型应用、人工智能量化和多因子模型，已发布多篇热度较高的相关深度报告，所在团队多次入围“机构投资者”最佳分析师团队。
摘要：
本次讲座将聚焦大模型与量化投资融合的前沿实践，系统解析大模型如何赋能量化投资。主讲人将结合华泰金工一线研究与实践经验，拆解大模型在因子挖掘、技术分析、文本选股增强、热点主题投资等关键场景的落地路径，分享实践应用的典型案例、技术框架与实操痛点。同时展望大模型与量化结合的未来及取舍之道，为同学们呈现大模型背景下的量化投研新范式。
北大经院工作坊第1241场
Reward Incentive and Moral Hazard: Informational Loss Sharing in Peer-to-peer Auto-Insurance
风险、保险与不确定性经济学工作坊
主讲人：黎韬（清华大学经济管理学院博士后）
主持人：
（人大财金）陈泽
（清华经管）冯润桓
（北大经院）贾若
参与老师：
（人大财金）魏丽、何林、胡文涛
（北大经院）郑伟
时间：2026年3月27日（周五）14:00-15:30
线上形式：腾讯会议
会议号：763 719 759
线下地点：中国人民大学教学二楼2203教室
主讲人简介：
黎韬，清华大学经济管理学院博士后，北美准精算师（ASA）候选（已通过所有考试）。黎韬于2020年在南开大学获得经济学学士学位，2025年在中国人民大学获得经济学博士学位。黎韬曾在The Geneva Papers on Risk and Insurance - Issues and Practice和《保险研究》等期刊上发表学术论文，曾获中国保险与风险管理国际年会（CICIRM）最佳论文奖、北美产险精算师协会（CAS）-腾讯天衍实验室数据分析大赛一等奖。主要研究兴趣为保险经济学与保险精算。
摘要：
Peer-to-peer(P2P) auto insurance is a novel risk-sharing mechanism that disrupts the traditional insurance model. Participants contribute initial deposits into special purpose accounts, and claims are settled weekly: indemnities are first paid out from the claimant’s own deposit, and any shortfall is then shared collectively among other participants. This paper theoretically analyzes the evolution of the moral hazard in a multi-period P2P insurance model, and empirically estimates its scale using a proprietary dataset from a Chinese P2P auto insurance platform. Our theoretical analysis reveals that participants with lower deposits tend to drive less cautiously, thereby exhibiting a higher likelihood of losses. This is because a greater portion of their indemnities is covered by others. Our empirical results validates the theoretical finding after controlling for alternative hypotheses (e.g. adverse selection), showing that 10% decrease in the deposit percentage leads to an approximate 8.2% increase in the weekly expected number of claims. We also indicate that moral hazards are more significant among female drivers of new vehicles.
北大经院工作坊第1242场
Can language models boost the power of randomized experiments without statistical bias?
计量、金融和大数据分析工作坊
主讲人：Xinwei Ma (Associate Professor of Economics at the University of California San Diego)
主持老师：（北大经院）王一鸣、巩爱博
参与老师：
（北大经院）刘蕴霆、王熙、王法、李少然
（北大国发院）黄卓、沈艳、张俊妮
时间：2026年3月27日（周五）10:00-11:30
地点：beat365中文官方网站101会议室
主讲人简介：
Xinwei Ma is an Associate Professor of Economics at the University of California San Diego. His research interests are interdisciplinary, focusing on developing robust statistical methodologies for economics, social sciences, and biomedical applications. Some specific research directions include kernel-based density estimation and falsification testing in regression discontinuity designs; semi-parametric estimation and inference with limited overlap; Mendelian randomization addressing weak instruments and the winner's curse; design and analysis of adaptive experiments; and language model-assisted inference and data routing.
Xinwei earned his Ph.D. in Economics from the University of Michigan in 2019. Prior to this, he received a Master of Finance from the University of Hong Kong, as well as a Master of Economics and a Bachelor of Science from Peking University. Xinwei serves as an Associate Editor for Econometric Theory, the Journal of Econometrics, and The Econometrics Journal. In 2025, he was elected a Fellow of the International Association for Applied Econometrics.
摘要：
Randomized experiments or randomized controlled trials (RCTs) are gold standards for causal inference, yet cost and sample-size constraints limit power. We introduce CALM (Causal Analysis leveraging Language Models), a statistical framework that integrates large language models (LLMs) generated insights of RCTs with established causal estimators to increase precision while preserving statistical validity. In particular, CALM treats LLM-generated outputs as auxiliary prognostic information and corrects their potential bias via a heterogeneous calibration step that residualizes and optimally reweights predictions. We prove that CALM remains consistent even when LLM predictions are biased and achieves efficiency gains over augmented inverse probability weighting estimators for various causal effects. In particular, CALM develops a few-shot variant that aggregates predictions across randomly sampled demonstration sets. The resulting U-statistic-like predictor restores i.i.d. structure and also mitigates prompt-selection variability. Empirically, in simulations calibrated to a mobile-app depression RCT, CALM delivers lower variance relative to other benchmarking methods, is effective in zero- and few-shot settings, and remains stable across prompt designs. By principled use of LLMs to harness unstructured data and external knowledge learned during pretraining, CALM provides a practical path to more precise causal analyses in RCTs.
北大经院工作坊第1243场
商业银行气候风险评估
生态、环境与气候变化经济学工作坊
主讲人：张俊杰（杜克大学尼古拉斯环境学院教授）
主持人：（北大经院）季曦
时间：2026年3月27日（周五）10:30-12:00
地点： beat365中文官方网站107会议室
主讲人简介：
张俊杰是杜克大学尼古拉斯环境学院教授、昆山杜克大学可持续投资研究项目主任，并担任上海金司南金融研究院院长及绿色金融60人论坛首席经济学家等社会职务；曾任加州大学圣地亚哥分校全球政策与战略学院助理教授、副教授，清华大学苏世民书院大众汽车可持续发展访问讲席教授。他的研究聚焦于环境经济学、绿色金融以及气候投融资等领域的实证问题。
摘要：
本研究阐述气候风险对商业银行信贷资产安全性的传导机制与量化评估挑战。实证分析表明，极端气温与降水等物理风险显著推高涉农信贷违约率，且存在显著的行业与作物异质性。研究同时指出，银行在管理转型风险时面临碳排放数据缺失、情景设计粗糙及风险货币化困难等结构性障碍。本研究以商业银行微观信贷数据为基础，为突破气候相关数据约束、开展微观审慎压力测试提供了方法论创新。这些发现为金融气候风险管理领域的学术研究提供了实证基础与框架参考。
供稿：科研与博士后办公室
美编：鱼尔
责编：度量、雨禾、雨田
