Zobrazit v EDS

Evaluating Dynamic Conditional Quantile Treatment Effects with Applications in Ridesharing

Uloženo v:
Podrobná bibliografie
Název: Evaluating Dynamic Conditional Quantile Treatment Effects with Applications in Ridesharing
Autoři: Li, Ting, Shi, Chengchun, Lu, Zhaohua, Li, Yi, Zhu, Hongtu
Zdroj: Journal of the American Statistical Association. 119:1736-1750
Publication Status: Preprint
Informace o vydavateli: Informa UK Limited, 2024.
Rok vydání: 2024
Témata: quantile treatment effect, spatialtemporal experiments, FOS: Computer and information sciences, Computer Science - Machine Learning, ridesourcing platform, 05 social sciences, Machine Learning (stat.ML), CCF-DiDi GAIA Collaborative Research Funds for Young Scholars and Program for Innovative Research Team of Shanghai University of Finance and Economics, policy evaluation, 01 natural sciences, Machine Learning (cs.LG), Methodology (stat.ME), varying coefficient models, EP/W014971/1, Statistics - Machine Learning, 0502 economics and business, 0101 mathematics, Li's research is partially supported by the Nation12101388, Statistics - Methodology, A/B testing
Popis: Many modern tech companies, such as Google, Uber, and Didi, use online experiments (also known as A/B testing) to evaluate new policies against existing ones. While most studies concentrate on average treatment effects, situations with skewed and heavy-tailed outcome distributions may benefit from alternative criteria, such as quantiles. However, assessing dynamic quantile treatment effects (QTE) remains a challenge, particularly when dealing with data from ride-sourcing platforms that involve sequential decision-making across time and space. In this article, we establish a formal framework to calculate QTE conditional on characteristics independent of the treatment. Under specific model assumptions, we demonstrate that the dynamic conditional QTE (CQTE) equals the sum of individual CQTEs across time, even though the conditional quantile of cumulative rewards may not necessarily equate to the sum of conditional quantiles of individual rewards. This crucial insight significantly streamlines the estimation and inference processes for our target causal estimand. We then introduce two varying coefficient decision process (VCDP) models and devise an innovative method to test the dynamic CQTE. Moreover, we expand our approach to accommodate data from spatiotemporal dependent experiments and examine both conditional quantile direct and indirect effects. To showcase the practical utility of our method, we apply it to three real-world datasets from a ride-sourcing platform. Theoretical findings and comprehensive simulation studies further substantiate our proposal. Supplementary materials for this article are available online Code implementing the proposed method is also available at: https://github.com/BIG-S2/CQSTVCM.
Druh dokumentu: Article
Jazyk: English
ISSN: 1537-274X
0162-1459
DOI: 10.1080/01621459.2024.2314316
DOI: 10.17615/gj41-bc93
DOI: 10.48550/arxiv.2305.10187
Přístupová URL adresa: http://arxiv.org/abs/2305.10187
http://eprints.lse.ac.uk/122488/
Rights: arXiv Non-Exclusive Distribution
Přístupové číslo: edsair.doi.dedup.....45cc6d386a6f5988fb7bbc809d997ec9
Databáze: OpenAIRE
Popis
Abstrakt:Many modern tech companies, such as Google, Uber, and Didi, use online experiments (also known as A/B testing) to evaluate new policies against existing ones. While most studies concentrate on average treatment effects, situations with skewed and heavy-tailed outcome distributions may benefit from alternative criteria, such as quantiles. However, assessing dynamic quantile treatment effects (QTE) remains a challenge, particularly when dealing with data from ride-sourcing platforms that involve sequential decision-making across time and space. In this article, we establish a formal framework to calculate QTE conditional on characteristics independent of the treatment. Under specific model assumptions, we demonstrate that the dynamic conditional QTE (CQTE) equals the sum of individual CQTEs across time, even though the conditional quantile of cumulative rewards may not necessarily equate to the sum of conditional quantiles of individual rewards. This crucial insight significantly streamlines the estimation and inference processes for our target causal estimand. We then introduce two varying coefficient decision process (VCDP) models and devise an innovative method to test the dynamic CQTE. Moreover, we expand our approach to accommodate data from spatiotemporal dependent experiments and examine both conditional quantile direct and indirect effects. To showcase the practical utility of our method, we apply it to three real-world datasets from a ride-sourcing platform. Theoretical findings and comprehensive simulation studies further substantiate our proposal. Supplementary materials for this article are available online Code implementing the proposed method is also available at: https://github.com/BIG-S2/CQSTVCM.
ISSN:1537274X
01621459
DOI:10.1080/01621459.2024.2314316