I'm running a fuzzy regression discontinuity in R. I'm trying to understand the interpretation of each coefficient in the regression output. I know that x is the effect of the running variable on the outcome to the left of the cutoff, x+x_right is the effect of the running variable on the outcome to the right of the cutoff, but I don't understand D and ins.
I know they must have to do with the effects of the instrument and the treatment, but I'm not exactly sure how.
My outcome y is binary, so I'm using a probit link function.
My code is this:
rdd_data_full = rdd_data(y=y,x=x,
cutpoint = cutpoint, z = z)
rdd_full = rdd_gen_reg(rdd_object=rdd_data_full,fun=glm,family=binomial(link='probit'),slope='separate',order=1)
And my output is this:
Call:
fun(formula = y ~ ., family = ..1, data = dat_step1, weights = weights)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.5845 -1.4928 0.8441 0.8807 0.9633
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.4827202 0.0152462 31.662 <2e-16 ***
D -0.1843083 0.0139786 -13.185 <2e-16 ***
x 0.0051875 0.0033124 1.566 0.117
x_right 0.0003811 0.0037638 0.101 0.919
ins 0.0184554 0.0196553 0.939 0.348
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 94258 on 75420 degrees of freedom
Residual deviance: 94076 on 75416 degrees of freedom
AIC: 94086
Number of Fisher Scoring iterations: 4
```
