1

With a RDD that has multiple cutoffs and fixed effects by individual, is there an R package that can be used to either 'normalise' or provide relevant propensity score weights? Bertanha "Regression Discontinuity Design with Many Thresholds" and Cattaneo “Interpreting Regression Discontinuity Designs with Multiple Cutoffs” have built on the theory, but is there an relevant R package to get stared. An example of the data:

Group   Name    P   C   T   P-C T*P-C  Value
A       DDT     15  10  1   5   5      110
A       DDT     13  10  1   3   3      123
A       MFC     18  10  1   8   8      133
A       MFC      9  10  0   -1  0      88
A       TFG      6  10  0   -4  0      79
A       TFG      6  10  0   -4  0      66
B       DDT     18  20  0   -2  0      98
B       DDT     17  20  0   -3  0      95
B       MFC     13  20  0   -7  0      94
B       MFC     25  20  1   5   5      122
B       TFG     23  20  1   3   3      143
B       TFG     23  20  1   3   3      141
C       DDT     28  30  0   -2  0      78
C       DDT     29  30  0   -1  0      88
C       MFC     33  30  1   3   3      134
C       MFC     37  30  1   7   7      123
C       TFG     33  30  1   3   3      133
C       TFG     34  30  1   4   4      164

The RD model being estimated has fixed effects (by 'Name') with clustered standard errors (by 'Name') plm1 <- plm(Value~I(P_C)*T, data = data, model = "within")

I'm looking to estimate the treatment effect across all thee groups (A, B, C), though they all have different cutoffs.

Any R packages that could suit this case?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.