A New Framework for Online Testing of Heterogeneous Treatment Effect


  • Miao Yu North Carolina State University
  • Wenbin Lu North Carolina State University
  • Rui Song North Carolina State University




We propose a new framework for online testing of heterogeneous treatment effects. The proposed test, named sequential score test (SST), is able to control type I error under continuous monitoring and detect multi-dimensional heterogeneous treatment effects. We provide an online p-value calculation for SST, making it convenient for continuous monitoring, and extend our tests to online multiple testing settings by controlling the false discovery rate. We examine the empirical performance of the proposed tests and compare them with a state-of-art online test, named mSPRT using simulations and a real data. The results show that our proposed test controls type I error at any time, has higher detection power and allows quick inference on online A/B testing.




How to Cite

Yu, M., Lu, W., & Song, R. (2020). A New Framework for Online Testing of Heterogeneous Treatment Effect. Proceedings of the AAAI Conference on Artificial Intelligence, 34(06), 10310-10317. https://doi.org/10.1609/aaai.v34i06.6594



AAAI Technical Track: Reasoning under Uncertainty