Hello everyone,

I'm testing premature deindustrialization as Rodrik (2015) I regress manufacture employment shares against log population, ln population squared, ln GDP per capita, and ln GDP per capita squared, country fixed effects and interactions terms for post-1990 of ln GDP and ln GDP squared.

After the regression, I plot the predicted values and I get a graph like this:

Array



However, I'm trying to get a smother graph such as Rodrik (2015) did (see below).

I would appreciate if you have any advice.

Enrique de la Rosa-R.

Array