Corpus ID: 211627842

Gaussian Process Regression Discontinuity

@inproceedings{Ornstein2020GaussianPR,
  title={Gaussian Process Regression Discontinuity},
  author={Joseph T. Ornstein and JBrandon Duck-Mayr},
  year={2020}
}
In applied settings, regression discontinuity (RD) designs often suffer from noisy data and low power. This tends to produce exaggerated causal effect estimates, typified by implausibly large slope and/or concavity parameters. We propose a new method for estimating causal effects in RD designs called Gaussian Process Regression Discontinuity (GPRD). This approach overcomes the major disadvantages of global polynomial estimators and does so with lower variance than local linear estimators. When… CONTINUE READING

Figures and Tables from this paper.

References

SHOWING 1-10 OF 24 REFERENCES
Regression Discontinuity Designs Using Covariates
  • 210
  • PDF
Beyond Power Calculations
  • A. Gelman, J. Carlin
  • Medicine
  • Perspectives on psychological science : a journal of the Association for Psychological Science
  • 2014
  • 436
  • PDF
Randomized experiments from non-random selection in U.S. House elections
  • 1,426
  • PDF
Gaussian Processes for Machine Learning
  • 12,854
  • PDF