Geographic Regression discontinuity

[Iddrisu Kambala]

Hello guys,
I am comparing the economic well-being of some group of people residing on opposite sides of a certain border. I believe there is a jump at the border, and I want to represent this graphically using Regression Discontinuity design. Is there anyone who has any experience in this, and would like to help ? Also I want to see how living conditions decline as one moves closer to the border.

[David Radwin]

I can't help directly but I can point you in the right direction. First, it seems you need a measure of distance from the border, where one side is denominated in positive numbers and the other side is denominated in negative numbers.

Second, it would help to have a program to do the RD analysis and graphs. For that, start with https://www.stata.com/meeting/columb...5_cattaneo.pdf and/or ssc describe rd. Click through the links and you will find citations to papers on geographic RD, too.

[David Benson]

You might also take a look at this post. The poster was looking to find the distance to the closest London subway station, but presumably you could modify it to measure distance to the border.

[Maxence Morlet]

Dear David Benson, apologies but the link in #3 seems to be dead. Please could you update it? It would be of great help if I could access that post. Thanks!