rdplot Y centre, c(0) p(2)
where I see a fall after cutoff, I have used different polynomials to check, and there is always fall after the cutoff.
However, when I am checking with rdrobust, I am getting a positive insignificant coefficient.
rdrobust Y centre, c(0) p(2) bwselect(msetwo) all
Mass points detected in the running variable.
Sharp RD estimates using local polynomial regression.
Cutoff c = 0 | Left of c Right of c Number of obs = 6868
-------------------+---------------------- BW type = msetwo
Number of obs | 2627 4241 Kernel = Triangular
Eff. Number of obs | 804 2043 VCE method = NN
Order est. (p) | 2 2
Order bias (q) | 3 3
BW est. (h) | 139.110 314.657
BW bias (b) | 217.590 406.097
rho (h/b) | 0.639 0.775
Unique obs | 13 33
Outcome: iron_folic_given1. Running variable: centre.
--------------------------------------------------------------------------------
Method | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------------+------------------------------------------------------------
Conventional | .00447 .07571 0.0590 0.953 -.14392 .152856
Bias-corrected | .02964 .07571 0.3915 0.695 -.11875 .178027
Robust | .02964 .09701 0.3055 0.760 -.160499 .219777
--------------------------------------------------------------------------------
Estimates adjusted for mass points in the running variable.
.
If only I use rdrobust with polynomial 1, i get a negative coefficient, but it is insignificant. Can anyone help me understand why?
where I see a fall after cutoff, I have used different polynomials to check, and there is always fall after the cutoff.
However, when I am checking with rdrobust, I am getting a positive insignificant coefficient.
rdrobust Y centre, c(0) p(2) bwselect(msetwo) all
Mass points detected in the running variable.
Sharp RD estimates using local polynomial regression.
Cutoff c = 0 | Left of c Right of c Number of obs = 6868
-------------------+---------------------- BW type = msetwo
Number of obs | 2627 4241 Kernel = Triangular
Eff. Number of obs | 804 2043 VCE method = NN
Order est. (p) | 2 2
Order bias (q) | 3 3
BW est. (h) | 139.110 314.657
BW bias (b) | 217.590 406.097
rho (h/b) | 0.639 0.775
Unique obs | 13 33
Outcome: iron_folic_given1. Running variable: centre.
--------------------------------------------------------------------------------
Method | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------------+------------------------------------------------------------
Conventional | .00447 .07571 0.0590 0.953 -.14392 .152856
Bias-corrected | .02964 .07571 0.3915 0.695 -.11875 .178027
Robust | .02964 .09701 0.3055 0.760 -.160499 .219777
--------------------------------------------------------------------------------
Estimates adjusted for mass points in the running variable.
.
If only I use rdrobust with polynomial 1, i get a negative coefficient, but it is insignificant. Can anyone help me understand why?
