This paper analyses regression discontinuity designs which can be used for the evaluation of development interventions. The specific focus of the investigation is the choice of the smoothing parameter that is used.
It "focus[es] on estimation by local linear regression, which was shown to be rate optimal (Porter, 2003). Investigation of an expected-squared-error-loss criterion reveals the need for regularization. [The authors] propose an optimal, data dependent, bandwidth choice rule. [They] illustrate the proposed bandwidth choice using data previously analyzed by Lee (2008), as well as in a simulation study based on this data set. The simulations suggest that the proposed rule performs well." (Imbens & Kalyanaraman, 2009)
Imbens, G., & Kalyanaraman, K. National Bureau of Economic Research, (2009). Optimal bandwidth choice for the regression discontinuity estimator(Working Paper 14726). Retrieved from website: http://www.nber.org/papers/w14726.pdf