司继春,上海财经大学博士,目前任教于上海对外经贸大学统计与信息学院,主要研究领域为微观计量经济学、产业组织理论,成果见诸 Journal of Business and Economic Statistics、《中国人口科学》、《系统工程理论与实践》等期刊。司老师专长于机器学习,尤其是基于机器学习的因果推断前沿方法,有多个大型数据分析项目的实战经验。业余时间里,司老师也经常在知乎上耐心作答,用通俗的语言普及统计和计量知识。他的知乎专栏名为「慧航」,关注者逾 31w,获赞超过 17w。他总能抽丝剥茧,把复杂的问题讲得清清楚楚。
张宏亮,美国麻省理工学院 (MIT) 博士,浙江大学经济学院新百人计划研究员,博士生导师。主要从事经济学微观实证研究,尤其偏爱因果推断方法在劳动经济学、公共经济学、发展经济学、城市经济学等领域的应用。研究成果见诸 International Economic Review (IER), Journal of European Economic Association (JEEA), Journal of Public Economics (JPubE), Journal of Development Economics (JDE, 2 篇), Journal of Urban Economics (JUE) 等专业领域顶级期刊。
C. 课程特色
懂原理、会应用。本次课程邀请了两位老师合作讲授,目的在于最大限度地实现理论与应用的有机结合。为期四天的课程,分成两个部分:第一部分讲解常用的机器学习算法和适用条件,以及文本分析和大语言模型;第二部分通过精讲 4-6 篇发表于 Top 期刊的论文,帮助大家理解各类机器学习算法的应用场景,以及它们与传统因果推断方法的巧妙结合。
以 Top 期刊论文为范例。目前多数人的困惑是不清楚如何将传统因果推断方法与机器学习结合起来。事实上,即便是 MIT 和 Harvard 的大牛们也都在「摸着石头过河」。为此,通过论文精讲和复现来学习这部分内容或许是目前最有效的方式了。张宏亮老师此前在浙江大学按照这一模式教授了「因果推断和机器学习」课程,效果甚佳:学生们能够逐渐建立起研究设计的理念,并在构造识别策略时适当地嵌入机器学习方法。
Deryugina, T., Heutel, G., Miller, N. H., Molitor, D., & Reif, J. (2019). The Mortality and Medical Costs of Air Pollution: Evidence from Changes in Wind Direction. American Economic Review, 109(12), 4178–4219. Link (rep), PDF, Appendix, Google, -cited-.
参考文献:
Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., Newey, W., & Robins, J. (2018). Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal, 21(1), C1–C68. Link, PDF, Google, Replication.
Gilchrist, D. S., & Sands, E. G. (2016). Something to talk about: Social spillovers in movie consumption. Journal of Political Economy, 124(5), 1339-1382. Link, PDF, Google, -cited-.
Knaus, M. C., Lechner, M., & Strittmatter, A. (2021). Machine learning estimation of heterogeneous causal effects: Empirical Monte Carlo evidence. The Econometrics Journal, 24(1), 134-161. Link, PDF, Google, -cited-.
Ahrens, A., Hansen, C. B., Schaffer, M. E., & Wiemann, T. (2024). ddml: Double/debiased machine learning in Stata. The Stata Journal, 24(1), 3–45. Link, PDF, Google.
Ahrens, A., Hansen, C. B., & Schaffer, M. E. (2023). pystacked: Stacking generalization and machine learning in Stata. The Stata Journal, 23(4), 909–931. Link, PDF, Google.
Anand, V., Bochkay, K., Chychyla, R., & Leone, A. (2020). Using Python for Text Analysis in Accounting Research. Foundations and Trends? in Accounting, 14(3–4), 128–359. Link, PDF, Google.
Benguria, F., Choi, J., Swenson, D. L., & Xu, M. J. (2022). Anxiety or pain? The impact of tariffs and uncertainty on Chinese firms in the trade war. Journal of International Economics, 103608. Link, PDF, Google.
Gentzkow, M., Kelly, B., & Taddy, M. (2019). Text as data. Journal of Economic Literature, 57(3), 535–574. Link, PDF, Appendix, Google, -cited-.
机器学习算法 (CART, LASSO, Random Forest, Gradient Boosting, Ensemble) 的介绍;
机器学习算法在实证预测案例中的应用;
Mullainathan & Spiess (2017, JEP) 的复现。
相关论文:
Kleinberg, J., Ludwig, J., Mullainathan, S., & Obermeyer, Z. (2015). Prediction Policy Problems. American Economic Review, 105(5), 491–495. Link (rep), PDF, Appendix, Google.
Mullainathan, S., & Spiess, J. (2017). Machine Learning: An Applied Econometric Approach. Journal of Economic Perspectives, 31(2), 87–106. Link (rep), PDF, Appendix, Google.
Einav, L., Finkelstein, A., Mullainathan, S., & Obermeyer, Z. (2018). Predictive Modeling of U.S. Health Care Spending in Late Life. Science, 360(6396), 1462–1465. Link, PDF, Google.
Kleinberg, J., Lakkaraju, H., Leskovec, J., Ludwig, J., & Mullainathan, S. (2018). Human Decisions and Machine Predictions. Quarterly Journal of Economics. Link (rep), PDF, Google.
Aiken, E., Bellue, S., Karlan, D., Udry, C., & Blumenstock, J. E. (2022). Machine Learning and Phone Data can Improve Targeting of Humanitarian Aid. Nature, 603(7903), 864–870. Link, PDF, Google.
在此基础上,他将进一步通过“美国职业安全与健康管理局”的随机检查对减少工作场所重大伤害发生率的现实案例(Johnson et al., AEJ:Applied, 2023)介绍机器学习辅助因果推断方法在优化政策干预对象选择、支持更有效的政策制定领域的应用:
最后,他将指导大家对 Johnson et al.(AEJ:Applied, 2023) 这篇文章的实证结果进行复现。
专题亮点:
因果森林算法;
基于因果森林算法的「异质性处置效应估算」;
基于机器学习的「目标干预对象选择」;
Johnson et al. (2023, AEJ) 的复现
相关论文:
Wager, S., & Athey, S. (2018). Estimation and Inference of Heterogeneous Treatment Effects using Random Forests. Journal of the American Statistical Association, 113(523), 1228–1242. Link, PDF, Google.
Davis, J. M. V., & Heller, S. B. (2020). Rethinking the Benefits of Youth Employment Programs: The Heterogeneous Effects of Summer Jobs. Review of Economics and Statistics, 102(4), 664–677. Link, PDF, Google.
Knittel, C. R., & Stolper, S. (2021). Machine Learning about Treatment Effect Heterogeneity: The Case of Household Energy Use. American Economic Review Papers and Proceedings, 111, 440–444. Link (rep), PDF, Appendix, Google.
Johnson, M. S., Levine, D. I., & Toffel, M. W. (2023). Improving Regulatory Effectiveness through Better Targeting: Evidence from OSHA. American Economic Journal: Applied Economics, 15(4), 30–67. Link (rep), PDF, Appendix, Google.
对于那些只对少数人产生影响的政策,如果不能将受影响的个体从样本中精准地区分出来,就导致评估结果失真、统计精度不足。为此,张宏亮老师将以 Cengiz et al. (JoLE,2022) 为例,介绍如何应用机器学习方法来 精准预测受政策影响的群体,从而实现对仅影响少数人的政策进行有效的因果效应评估。在这篇论文中,Cengiz et al.(2022) 首先通过机器学习方法建立了一个预测模型,预测出个体劳动者受最低工资上调影响的概率,进而挑选出受最低工资上调影响概率最高的 10% 的劳动者,最后结合事件研究法,分析「最低工资上调事件」对他们的工资水平和就业的影响。
专题亮点:
机器学习在高维模型选择中的应用;
PDS Lasso 方法及其应用;
机器学习精准预测受政策影响的群体;
如何评估只对少数人有影响的政策?
JEP (2014) 文章实证结果的复制。
相关论文:
Belloni, A., Chernozhukov, V., & Hansen, C. (2013). Inference on Treatment Effects after Selection among High-Dimensional Controls. Review of Economic Studies, 81(2), 608–650. Link (rep), PDF, Google.
Belloni, A., Chernozhukov, V., & Hansen, C. (2014). High-Dimensional Methods and Inference on Structural and Treatment Effects. Journal of Economic Perspectives, 28(2), 29–50. Link (rep), PDF, Appendix, Google.
Lowes, S., & Montero, E. (2021). The Legacy of Colonial Medicine in Central Africa. American Economic Review, 111(4), 1284–1314. Link (rep), PDF, Appendix, Google.
Angrist, J. D., & Frandsen, B. (2022). Machine Labor. Journal of Labor Economics, 40(S1), S97–S140. Link (rep), PDF, Appendix, Google.
Cengiz, D., Dube, A., Lindner, A., & Zentler-Munro, D. (2022). Seeing beyond the Trees: Using Machine Learning to Estimate the Impact of Minimum Wages on Labor Market Outcomes. Journal of Labor Economics, 40(S1), S203–S247. Link(rep),PDF, Google.