EconBase
← All papers

Combining Observational and Experimental Data to Improve Efficiency Using Imperfect Instruments

George Z. Gui

arXiv 10 Oct 2020 · Econometrics · publishedMarketing Science (2023) · 5 citations (OpenAlex)

arXiv:2010.05117 · PDF · DOI · OpenAlex · Extracted main text

Abstract

Randomized controlled trials generate experimental variation that can credibly identify causal effects, but often suffer from limited scale, while observational datasets are large, but often violate desired identification assumptions. To improve estimation efficiency, I propose a method that leverages imperfect instruments - pretreatment covariates that satisfy the relevance condition but may violate the exclusion restriction. I show that these imperfect instruments can be used to derive moment restrictions that, in combination with the experimental data, improve estimation efficiency. I outline estimators for implementing this strategy, and show that my methods can reduce variance by up to 50%; therefore, only half of the experimental sample is required to attain the same statistical precision. I apply my method to a search listing dataset from Expedia that studies the causal effect of search rankings on clicks, and show that the method can substantially improve the precision.

Citation extraction

22
references
49
in-text mentions
22
distinct cited
0
self-citations
9,174
main-text words

appendix boundary found by appendix_command · 66% of the source is main text. Read the extracted text to check this.

Most heavily cited references

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.

ReferenceIntensityMentionsSectionsMain text
1Ursu, R. M (2018) The Power of Rankings: Quantifying the Effect of Rankings on Online Consumer Search and Purchase Decisions1.00073100%
2Lewis, R. A. and Rao, J. M (2015) The Unfavorable Economics of Measuring the Returns to Advertising*0.81142100%
3Imbens, G. W. and Lancaster, T (1994) Combining Micro and Macro Data in Microeconometric Models0.73732100%
4Li, L., Chen, S., Kleban, J., and Gupta, A (2015) Counterfactual Estimation and Optimization of Click Metrics in Search Engines: A Case Study0.64422100%
5Nevo, A. and Rosen, A. M (2010) Identification With Imperfect Instruments0.64422100%
6Wang, X., Bendersky, M., Metzler, D., and Najork, M (2016) Learning to Rank with Selection Bias in Personal Search0.64422100%
7Deng, A., Xu, Y., Kohavi, R., and Walker, T (2013) Improving the sensitivity of online controlled experiments by utilizing pre-experiment data0.58531100%
8Peysakhovich, A. and Lada, A (2016) Combining observational and experimental data to find heterogeneous treatment effects0.58531100%
9Bertrand, M., Karlan, D., Mullainathan, S., Shafir, E., and Zinman, J (2010) What's Advertising Content Worth? Evidence from a Consumer Credit Marketing Field Experiment*0.5114225%
10Chen, M. K., Chevalier, J. A., Rossi, P. E., and Oehlsen, E (2019) The Value of Flexible Work: Evidence from Uber Drivers0.5113233%

Showing the top 10 of 22 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
1The Identification Power of Combining Experimental and Observational Data for Distributional Treatment Effect Parameters0.40511
2Cross-Validated Causal Inference: a Modern Method to Combine Experimental and Observational Data0.40511