QJE
Human Decisions and Machine Predictions
Counterfactual Reasoning and Learning Systems: The Example of Computational Advertising
https://jmlr.org/papers/volume14/bottou13a/bottou13a.pdf
因果推断领域:
1. Judea Pearl 的 do-calculus,用概率图模型进行因果推断。
Pearl, J. (2012). The do-calculus revisited. In Proceedings of the Twenty-Eighth Conference on Uncertainty in Artificial Intelligence (pp. 3-11).
2. Johansson et al (2016) 用 representation learning进行 Rubin 框架下的反事实推断。
Johansson, F., Shalit, U., & Sontag, D. (2016). Learning representations for counterfactual inference. In
International conference on machine learning (pp. 3020-3029). PMLR.
3. Jason Hartford et al (2017) 的 DeepIV,用deep learning 来寻找工具变量进行反事实推断研究。
Jason Hartford, Greg Lewis, Kevin Leyton-Brown, and Matt Taddy. Deep IV: A flexible approach for counterfactual prediction. Proceedings of the 34th International Conference on Machine Learning, 2017.
4. Susan Athey 的 causal tree 模型,用树模型进行因果推断。
Athey, S., & Imbens, G. (2016). Recursive partitioning for heterogeneous causal effects.
Proceedings of the National Academy of Sciences, 113 (27), 7353-7360.
行为经济学领域:
1. Param Singh et al (2011) 用了 hidden markov model,来研究用户的隐含状态对可观测行为的影响。
Singh, P. V., Tan, Y., & Youn, N. (2011). A hidden Markov model of developer learning dynamics in open source software projects. Information Systems Research, 22(4), 790-807.
2. Jacobs et al (2016) 用了 Topic Modeling,来预测用户在线购买商品的行为。
Jacobs, B. J., Donkers, B., & Fok, D. (2016). Model-based purchase predictions for large assortments. Marketing Science,35(3), 389-404.
3. Peter Fader 的 Pareto/NBD model,用统计学模型预测和估计顾客的终身价值。
Fader, P. S., Hardie, B. G., & Lee, K. L. (2005). “Counting your customers” the easy way: An alternative to the Pareto/NBD model. Marketing science,24 (2), 275-284.
4. Peter Hoff 的 Latent Space Model,用统计学模型估计网络结构上用户的传播信息行为。
Hoff, P. D., Raftery, A. E., & Handcock, M. S. (2002). Latent space approaches to social network analysis.
5. Bayesian Learning,很多经济学、金融学、营销学的TOP论文都讨论过类似于 Customer Learning, Investor Learning 的问题。
Chance, D. M., Hillebrand, E., & Hilliard, J. E. (2008). Pricing an option on revenue from an innovation: An application to movie box office revenue. Management Science,54(5), 1015-1028.
Narayan, V., Rao, V. R., & Saunders, C. (2011). How peer influence affects
attribute preferences: A Bayesian updating mechanism.Marketing Science,30(2), 368-384.
Kwon, H. D., & Lippman, S. A. (2011). Acquisition of project-specific assets with Bayesian updating.
Operations research,59(5), 1119-1130.
博弈论领域:
Reinforcement Learning 和 GAN天然就是用来研究博弈论的,比如 Igami 教授论述了动态离散选择模型与 Reinforcement Learning 与我们已知的一些人工智能算法间的关系。这个方向在工程学里一直有人研究。像什么演化博弈、Agent model 等等
Igami, M. (2020). Artificial intelligence as structural
estimation: Deep Blue, Bonanza, andAlphaGo.The Econometrics Journal,23(3), S1-S24.
Charpentier, A., Elie, R., & Remlinger, C. (2021). Reinforcement learning in economics and finance.
Computational Economics, 1-38.
挖掘新的自变量:
1. Netzer et al (2012) 用 text ming 的方法,挖掘厂商之间的竞争关系是市场份额
Netzer, O., Feldman, R., Goldenberg, J., & Fresko, M. (2012). Mine your own business: Market-structure surveillance through text mining. Marketing Science,31(3), 521-543.
2.Zhang et al (2021) 用 image mining的方法,研究 Airbnb上房间图片质量与用户需求之间的关系。
Zhang, S., Lee, D., Singh, P. V., & Srinivasan, K. (2021). What makes a good image? Airbnb demand analytics leveraging interpretable image features.Management Science.
3.Liu et al (2019) 用 video mining 的方法,研究 youtube 视频对于患者康复的影响
Liu, X., Zhang, B., Susarla, A., & Padman, R. (2019). Go to YouTube and call me in the morning: use of social media for chronic conditions.MIS Quarterly.
业界正在做的项目:
google 的 What-If 项目: what-if-tool
很多大厂的推荐算法:讨论推荐算法是否公平的问题, fairness of recommendation algorithms
Reference
https://www.zhihu.com/question/37870042