Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
149 tokens/sec
GPT-4o
9 tokens/sec
Gemini 2.5 Pro Pro
47 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Identifying Treatment and Spillover Effects Using Exposure Contrasts (2403.08183v3)

Published 13 Mar 2024 in econ.EM and stat.ME

Abstract: To report spillover effects, a common practice is to regress outcomes on statistics capturing treatment variation among neighboring units. This paper studies the causal interpretation of nonparametric analogs of these estimands, which we refer to as exposure contrasts. We demonstrate that their signs can be inconsistent with those of the unit-level effects of interest even under unconfounded assignment. We then provide interpretable restrictions under which exposure contrasts are sign preserving and therefore have causal interpretations. We discuss the implications of our results for cluster-randomized trials, network experiments, and observational settings with peer effects in selection into treatment.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (30)
  1. Aronow, P. and C. Samii, “Estimating Average Causal Effects Under General Interference, with Application to a Social Network Experiment,” Annals of Applied Statistics, 2017, 11 (4), 1912–1947.
  2. Auerbach, E. and M. Tabord-Meehan, “The Local Approach to Causal Inference Under Network Interference,” arXiv preprint arXiv:2105.03810, 2023.
  3.   , J. Auerbach, and M. Tabord-Meehan, “Discussion of “Causal Inference with Misspecified Exposure Mappings: Separating Definitions and Assumptions”,” Biometrika, 2024, 111 (1), 21–24.
  4. Bajari, P., H. Hong, and S. Ryan, “Identification and Estimation of a Discrete Game of Complete Information,” Econometrica, 2010, 78 (5), 1529–1568.
  5.   ,   , J. Krainer, and D. Nekipelov, “Estimating Static Models of Strategic Interactions,” Journal of Business and Economic Statistics, 2010, 28 (4), 469–482.
  6. Balat, J. and S. Han, “Multiple Treatments with Strategic Substitutes,” Journal of Econometrics, 2023, 234 (2), 732–757.
  7. Blandhol, C., J. Bonney, M. Mogstad, and A. Torgovitsky, “When is TSLS Actually LATE?,” NBER: working paper 29709, 2022.
  8. Bugni, F., I. Canay, and S. McBride, “Decomposition and Interpretation of Treatment Effects in Settings with Delayed Outcomes,” arXiv preprint arXiv:2302.11505, 2023.
  9. Chetverikov, D., J. Hahn, Z. Liao, and S. Sheng, “Logit-Based Alternatives to Two-Stage Least Squares,” arXiv preprint arXiv:2312.10333, 2023.
  10. de Chaisemartin, C. and X. d’Haultfoeuille, “Two-Way Fixed Effects Estimators with Heterogeneous Treatment Effects,” American Economic Review, 2020, 110 (9), 2964–2996.
  11. Eck, D., O. Morozova, and F. Crawford, “Randomization for the Direct Effect of an Infectious Disease Intervention in a Clustered Study Population,” arXiv preprint arXiv:1808.05593, 2018.
  12. Forastiere, L., E. Airoldi, and F. Mealli, “Identification and Estimation of Treatment and Interference Effects in Observational Studies on Networks,” Journal of the American Statistical Association, 2021, 116 (534), 901–918.
  13. Goldsmith-Pinkham, P., P. Hull, and M. Kolesár, “Contamination Bias in Linear Regressions,” National Bureau of Economic Research working paper, 2024.
  14. Harshaw, C., F. Sävje, and Y. Wang, “A Design-Based Riesz Representation Framework for Randomized Experiments,” arXiv preprint arXiv:2210.08698, 2022.
  15. Heckman, J. and R. Pinto, “Econometric Causality: The Central Role of Thought Experiments,” NBER working paper, 2023.
  16. Hudgens, M. and E. Halloran, “Toward Causal Inference with Interference,” Journal of the American Statistical Association, 2008, 103 (482), 832–842.
  17. Imbens, G. and J. Angrist, “Identification and Estimation of Local Average Treatment Effects,” Econometrica, 1994, pp. 467–475.
  18. Leung, M., “Causal Inference Under Approximate Neighborhood Interference,” arXiv preprint arXiv:1911.07085v1, 2019.
  19.   , “Causal Inference Under Approximate Neighborhood Interference,” Econometrica, 2022, 90 (1), 267–293.
  20.    and P. Loupos, “Unconfoundedness with Network Interference,” arXiv preprint arXiv:2211.07823, 2023.
  21. Lin, Z. and F. Vella, “Selection and Endogenous Treatment Models with Social Interactions: An Application to the Impact of Exercise on Self-Esteem,” IZA DP No. 14167, 2021.
  22.    and H. Xu, “Estimation of Social-Influence-Dependent Peer Pressure in a Large Network Game,” The Econometrics Journal, 2017, 20 (3), S86–S102.
  23. Manski, C., “Identification of Treatment Response with Social Interactions,” The Econometrics Journal, 2013, 16 (1), S1–S23.
  24. Ogburn, E., O. Sofrygin, I. Diaz, and M. van der Laan, “Causal Inference for Social Network Data,” arXiv preprint arXiv:1705.08527, 2022.
  25. Pearl, J., “Direct and Indirect Effects,” in “Probabilistic and Causal Inference: The Works of Judea Pearl” 2022, pp. 373–392.
  26. Sävje, F., “Causal Inference with Misspecified Exposure Mappings,” Biometrika, 2024, 111 (1), 1–15.
  27.   , “Rejoinder: Causal Inference with Misspecified Exposure Mappings: Separating Definitions and Assumptions,” Biometrika, 2024, 111 (1), 25–29.
  28. VanderWeele, T. and E. Tchetgen, “Effect Partitioning Under Interference in Two-Stage Randomized Vaccine Trials,” Statistics and Probability Letters, 2011, 81 (7), 861–869.
  29. Vytlacil, E., “Independence, Monotonicity, and Latent Index Models: An Equivalence Result,” Econometrica, 2002, 70 (1), 331–341.
  30. Xu, H., “Social Interactions in Large Networks: A Game Theoretic Approach,” International Economic Review, 2018, 59 (1), 257–284.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now