Difference between parametric vs nonparametric Regression Discontinuity Design

[Jun Park]

Hello, I'm trying to run an RDD exploiting a close election (where the margin of votes was less than 5%) in Brazil.
However, I am having trouble understanding the difference between parametric and nonparametric Regression Discontinuity Design. Can anyone help me out with grappling with the different concepts? Thank you!

[George Ford]

Nonparametric imposes no functional form (linear) and will use locally-weighted regression or some such.

You might look at something like this to get a feel for it:

twoway lfit var1 time || lowess var1 time

This might be helpful.

https://www.survey.mdrc.org/sites/default/files/RDD%20Guide_Full%20rev%202016_0.pdf