What is spatial regression discontinuity?
Spatial regression discontinuity is a special case that recognizes geographic borders as sharp cutoff points. Geographic distance represents the assignment variable; treatment-defining border represents the cutoff. • Two-dimensional space (e.g., latitude and longitude) must be reduced to one one-dimensional distance.
Why high order polynomials should not be used in regression discontinuity designs?
We argue that controlling for global high-order polynomials in regression discontinuity analysis is a flawed approach with three major problems: it leads to noisy estimates, sensitivity to the degree of the polynomial, and poor coverage of confidence intervals.
When to use difference in difference vs discontinuity?
When you take the first difference of the outcome for each group over time, the time-invariant effect is subtracted out and doesn’t contaminate the comparison in the second difference. So RD requires different assumptions and less data that DID, but it estimates a more local effect around the cutoff.
What do you need for fuzzy regression discontinuity?
For regression discontinuity you need a continuous “score” variable that determined who got the treatment. The score variable preferably has a clear cutoff that determines assignment. If there is no clear cutoff, you may still be able to use it as an instrumental variable for what’s known as a “fuzzy” regression discontinuity.
Is the regression model equal to the state fixed effect?
regression model is actually equal to we can simply take the expectations of the regression equation and calculate the difference: We can easily see that 1. the Diff-in-Diff estimator is equal to the coefficient of the interaction and 2. that the first difference takes out the state fixed effect and taking the difference of the
What’s the difference between Rd and did data?
So RD requires different assumptions and less data that DID, but it estimates a more local effect around the cutoff. DID requires panel data and is more global in some sense.