Optimal bandwidth estimated using a common mse optimal bandwidth selector based on calonico et al. We focus on estimation by local linear regression, which was shown to be rate optimal porter, 2003. Optimal bandwidth choice for the regression discontinuity estimator guido imbens, karthik kalyanaraman. Identification and estimation of treatment effects with a regression discontinuity design. I am having issues with the estimation of an optimal bandwidth using the rd command i am using stata 14. We describe a major upgrade to the stata and r rdrobust package. The default bandwidth from imbens and kalyanaraman 2009 is designed to minimize mse. R code to implement the imbenskalyanaraman bandwidth selection in rdd.
The most popular choices of kernel are the uniform kernel and the triangular kernel, which give equal weighting and linear downweighting to the observations with x i. Optimal bandwidth rd estimator by fujiimbenskalyanaraman, as. Software for implementing these methods is available in matlab, stata, and r. Investigation of an expectedsquared errorloss criterion reveals the need for regularization. Further details and exact formulas are given in the sa to conserve space. Mse optimal bandwidth choice yields an mse optimal rd point estimator, but is by construction invalid for inference.
Under stata versions 10 or later using lpoly to construct local regression estimates. Optimal bandwidth choice for robust biascorrected inference in regression discontinuity designs. January 2010 abstract we investigate the problem of optimal choice of the smoothing parameter bandwidth for the regression discontinuity estimator. The main new features of this upgraded version are as follows. Ikbandwidth imbenskalyanaraman optimal bandwidth calculation description ikbandwidthcalculates the imbenskalyanaraman optimal bandwidth for local linear regression in regression discontinuity designs.
Dear statalist members, i would like to use the rdob optimal bandwidth rd estimator by fuji imbens kalyanaraman, as. Software for regressiondiscontinuity designs matias d. Methods for constructing and assessing propensity scores. If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Rbc con dence intervals and related inference procedures remain valid even when the mse optimal bandwidth is used calonico, cat. Following david lees pioneering work, numerous scholars have applied the regression discontinuity rd design to popular elections. The stata newsa periodic publication containing articles on using stata and tips on using the software, announcements of new releases and updates, feature highlights, and other announcements of interest to interest to stata usersis sent to all stata users and those who request information about stata from us. Usage ikbandwidthx, y, cutpoint null, verbose false, kernel triangular arguments x a numerical vector which is the running variable. Citations of optimal bandwidth choice for the regression. If, on the other hand, you choose to wide a bandwidth, the resulting graph is \oversmoothed, which means it misses some of the most important features of the density figure 8 kernel density estimation with di erent bandwidths 0. Imbens and kalyanaraman 2012 optimal bandwidth choice for the regression discontinuity estimator restud other methods existsee lee and lemieux 2010 p. All of the papers that use local linear regressions also use a type of standard procedure to choose the optimal bandwidth either imbens and lemieux 2008 or imbens and kalyanaraman 2011. Rdestimate supports both sharp and fuzzy rdd utilizing the aer package for 2sls regression under the fuzzy design. Multidimensional regression discontinuity and regression kink designs with differenceindifferences.
See stata and matlab code here code from imbens software page. The optimal bandwidth will tend to be larger for a fuzzy design due to the. Multidimensional regression discontinuity and regression. Jul 15, 2011 the sensitivity of bandwidth to scale is particularly undesirable, but also serves to illustrate what i have said elsewhere. Rd designs can be invalid if individuals can precisely manipulate the assignment variable. The two differ by a constant as explained in the wp. Furthermore, we provide optimal bandwidth selectors. We investigate the choice of the bandwidth for the regression discontinuity estimator. Robust datadriven inference in the regressiondiscontinuity. Graphical presentation of regression discontinuity results. R code to implement the imbens kalyanaraman bandwidth selection in rdd. We focus on estimation by local linear regression, which was shown to have attractive properties porter, j. This is some work i did one weekend 20120617 to reconcile the estimates of optimal bandwidth provided by code written by devin caughey and code provided on guido imbens website. Regression discontinuity designs in economics 283 assigned to individuals or units with a value of x greater than or equal to a cutoff value c.
Reemployment probabilities over the business cycle. Optimal bandwidth choice for robust bias corrected. Using a simulation design that is based on empirical data, a recent study by huber et al. Ikbandwidth imbens kalyanaraman optimal bandwidth calculation description ikbandwidthcalculates the imbens kalyanaraman optimal bandwidth for local linear regression in regression discontinuity designs. Matlab and stata software for implementing this bandwidth rule is available on the web site economics. The stata package rdrobust accompanying calonico et al. This mse optimal bandwidth choice yields a mse optimal rd point estimator, but is by construction invalid for inference. Optimal bandwidth choice for robust biascorrected inference. Optimal datadriven regression discontinuity plots, with sebastian calonico and rocio titiunik. Optimal bandwidth for rd nber working paper series optimal. The kernel and bandwidth serve to localize the regression fit near the cutoff. Notes for matlab and stata regression discontinuity software.
Optimal bandwidth selection for the fuzzy regression. In the end, i decided to replace the poisson with a dummy. Optimal bandwidth for rd nber working paper series. Implementing matching estimators for average treatment effects in stata. R code to implement the imbenskalyanaraman bandwidth. Contrary to the assumptions of rd, however, we show that bare winners and bare losers in u. In this article, we describe a major upgrade to the stata and r software package. Robust bias corrected rbc inference methods provide a natural solution to this problem.
I tried in stata to write a crossvalidation code based on imbens and lemieux 2008, but i was not able to replicate any other additional bandwidth. Optimal bandwidth choice for the regression discontinuity. Simply select your manager software from the list below and click on download. Since the focus is solely on the change in the value of the regression function at the threshold, standard plugin methods and crossvalidation methods, which choose a bandwidth that is optimal for estimating the regression function over the entire support, do not yield an optimal bandwidth here.
Optimal rd bandwidth choice also for rectangular kernel. We describe a major upgrade to the stata and r rdrobust package, which provides a wide array of estimation. First, we present rdrobust, a command that implements the robust biascorrected confidence intervals proposed in calonico, cattaneo, and titiunik 2014d, econometrica 82. We investigate the problem of optimal choice of the smoothing parameter bandwidth for the regression discontinuity estimator. Nov 12, 2019 the kernel and bandwidth serve to localize the regression fit near the cutoff. Matching estimators implementing matching estimators for average treatment effects in stata stata 8 readme. This is some work i did one weekend 20120617 to reconcile the estimates of optimal bandwidth provided by code written by devin caughey and code provided on guido imbens website first, lets get some test data. Mar 10, 2015 all of the papers that use local linear regressions also use a type of standard procedure to choose the optimal bandwidth either imbens and lemieux 2008 or imbens and kalyanaraman 2011. Optimal bandwidth choice for the regression discontinuity estimator. If there are thresholds whereby some observations receive the treatment above it, other those below it do not, and those immediately above or below that threshold are similar, we can use the difference of the outcome between those just above and those just below the threshold to estimate the causal effect of the treatment. C14 abstract we investigate the problem of optimal choice of the smoothing parameter bandwidth for the regression discontinuity estimator. Dec 16, 2015 identification and estimation of treatment effects with a regression discontinuity design.
Mseoptimal bandwidth choice yields an mseoptimal rd point estimator, but is by. The paper is highly technical, but they have generously provided software to implement their bandwidth selection process in both matlab and stata. Rbc confidence intervals and related inference procedures remain valid even when the mse optimal bandwidth is used calonico, cattaneo. Optimal bandwidth selection for the fuzzy regression discontinuity estimator yoichi arai a and hidehiko ichimurab anational graduate institute for policy studies grips, 7221 roppongi, minatoku, tokyo 1068677, japan. The choice of bandwidth, h, is the key parameter when implementing the rd estimator, and we discuss this choice in detail below. Nber working paper series optimal bandwidth choice for the regression discontinuity estimator guido imbens karthik kalyanaraman working paper 14726 national bureau of economic research 1050 massachusetts avenue cambridge, ma 028 february 2009 financial support for this research was generously provided through nsf grants 0452590 and 0820361. Minimum detectable effect size computations for cluster. Optimal bandwidth estimated using a common mseoptimal bandwidth selector based on calonico et al. Robust datadriven inference in the regressiondiscontinuity design sebastian calonico, matias d. Regression discontinuity design in stata part 1 stata daily. Regression discontinuity issue with optimal bandwidth. Local linear regressions are performed to either side of the cutpoint using the imbenskalyanaraman optimal bandwidth calculation, 0. Imbens and rubin 2015 and abadie and cattaneo 2018, and references therein.
Optimal bandwidth choice for the regression discontinuity estimator guido imbens and karthik kalyanaraman nber working paper no. Using 1, the mseoptimal bandwidth choice for the rd treatment e ect estimator. Local linear regressions are performed to either side of the cutpoint using the imbens kalyanaraman optimal bandwidth calculation, ikbandwidth. The optimal bandwidth authors matthieu stigler references. Regression discontinuity design in stata part 1 stata. In this article, we introduce three commands to conduct robust datadriven statistical inference in regressiondiscontinuity rd designs.
Optimal bandwidth considerations apply to datadriven approaches when the outcome variable is available in advance. The sensitivity of bandwidth to scale is particularly undesirable, but also serves to illustrate what i have said elsewhere. Optimal bandwidth choice for robust bias corrected inference. The infeasible mseoptimal bandwidth choice h mse can be used to construct an mse.
1155 1616 1312 601 56 1107 819 932 656 578 149 640 1480 949 1151 1336 853 845 745 1457 1578 1461 102 391 840 1320 378 11 90 669 794 256 38 249 1642 1260 812 154 545 570 1484 1016 875 680 555 875 192 581 1322 73 1237