<p><p>We investigate the problem of optimal choice of the smoothing parameter (bandwidth) for the regression discontinuity estimator. We focus on estimation by local linear regression, which was shown to be rate optimal (Porter, 2003). We derive the optimal bandwidth. This optimal bandwidth depends on unknown functionals of the distribution of the data and we propose specific, consistent, estimators for these functionals to obtain a fully data-driven bandwidth choice that has the "asymptotic no-regret" property. We illustrate our proposed bandwidth, and the sensitivity to the choices made in this bandwidth proposal, using a data set previously analyzed by Lee (2008), as well as a small simulation study based on the Lee data set. The simulations suggest that the proposed rule performs well.</p></p>
Authors
Stanford Graduate School of Business
American Enterprise Institute
Working Paper details
- DOI
- 10.1920/wp.cem.2010.0510
- Publisher
- IFS
Suggested citation
Imbens, G and Kalyanaraman, K. (2010). Optimal bandwidth choice for the regression discontinuity estimator. London: IFS. Available at: https://ifs.org.uk/publications/optimal-bandwidth-choice-regression-discontinuity-estimator (accessed: 18 May 2024).
Related documents
More from IFS
Understand this issue
Where next for the state pension?
13 December 2023
Social mobility and wealth
12 December 2023
Autumn Statement 2023: IFS analysis
23 November 2023
Policy analysis
The past and future of UK health spending
14 May 2024
Recent trends in and the outlook for health-related benefits
19 April 2024
Progression of nurses within the NHS
12 April 2024
Academic research
Keeping the peace whilst getting your way: Information, persuasion and intimate partner violence
17 May 2024
The role of hospital networks in individual mortality
13 May 2024
Forced displacement, mental health, and child development: Evidence from Rohingya refugees
10 May 2024