Ruoxuan Xiong, Alex Chin, Sean J. Taylor
arXiv 10 Jun 2024 · Statistics — Methodology · 3 citations (OpenAlex)
arXiv:2406.06768 · PDF · DOI · OpenAlex · Extracted main text
We study the design and analysis of switchback experiments conducted on a single aggregate unit. The design problem is to partition the continuous time space into intervals and switch treatments between intervals, in order to minimize the estimation error of the treatment effect. We show that the estimation error depends on four factors: carryover effects, periodicity, serially correlated outcomes, and impacts from simultaneous experiments. We derive a rigorous bias-variance decomposition and show the tradeoffs of the estimation error from these factors. The decomposition provides three new insights in choosing a design: First, balancing the periodicity between treated and control intervals reduces the variance; second, switching less frequently reduces the bias from carryover effects while increasing the variance from correlated outcomes, and vice versa; third, randomizing interval start and end points reduces both bias and variance from simultaneous experiments. Combining these insights, we propose a new empirical Bayes design approach. This approach uses prior data and experiments for designing future experiments. We illustrate this approach using real data from a ride-sharing platform, yielding a design that reduces MSE by 33% compared to the status quo design used on the platform.
appendix boundary found by none_found · 100% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Bojinov, Iavor, David Simchi-Levi, Jinglong Zhao (2023) Design and analysis of switchback experiments | 1.000 | 9 | 4 | 100% |
| 2 | Hu, Yuchen, Stefan Wager (2022) Switchback experiments under geometric mixing | 1.000 | 6 | 4 | 100% |
| 3 | Masoero, Lorenzo, Guido Imbens, Thomas Richardson, James McQueen, Su… (2023) Efficient switchback experiments via multiple randomization designs | 0.843 | 3 | 3 | 100% |
| 4 | Xiong, Ruoxuan, Susan Athey, Mohsen Bayati, Guido W Imbens (2023) Optimal experimental design for staggered rollouts self | 0.843 | 3 | 3 | 100% |
| 5 | Basse, Guillaume W, Yi Ding, Panos Toulis (2023) Minimax designs for causal effects in temporal experiments with treatment habituation | 0.737 | 3 | 2 | 100% |
| 6 | Wu, Yuhang, Zeyu Zheng, Guangyu Zhang, Zuohua Zhang, Chu Wang (2022) Non-stationary a/b tests | 0.644 | 2 | 2 | 100% |
| 7 | Chamandy, Nicholas (2016) Experimentation in a ridesharing marketplace | 0.511 | 2 | 1 | 100% |
| 8 | Wu, CF Jeff, Michael S Hamada (2011) Experiments: planning, analysis, and optimization\/ | 0.511 | 2 | 1 | 100% |
| 9 | Cooprider, Joe, Shima Nassiri (2023) The science of price experiments in the amazon store | 0.405 | 1 | 1 | 100% |
| 10 | Kastelman, David, Raghav Ramesh (2018) Switchback tests and randomized experimentation under network effects at doordash | 0.405 | 1 | 1 | 100% |
Showing the top 10 of 59 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | Calibrated Horizon-Weighted Local Projection Designs for Markov Switchbacks | 0.644 | 2 | 2 |
| 2 | Can We Validate Counterfactual Estimations in the Presence of General Network Interference? | 0.405 | 1 | 1 |
| 3 | Experimental Design for Matching | 0.405 | 1 | 1 |