Interpretation of coefficients for continuous variables using PPML

[Stan Peterburgsky]

Hello,

I would like to make sure I understand how to interpret coefficients for continuous variables using PPML.
I have the following results (dependent variable is migration, not logged):

		(1)
log (Population origin)		0.8992***
		(19.873)
log (Population destination)		0.7502***
		(12.958)
log (Distance)		0.0473
		(0.661)
Contiguity		0.5465***
		(5.631)
Common language		1.2386***
		(11.444)
log (X origin)		-0.2745
		(-1.512)
log (X destination)		-0.5573***
		(-3.566)

Would it be correct to say that a 1% increase in "X origin" is associated with a -0.2745% change in migration and, likewise, a 1% increase in "X destination" is associated with a -0.5573% change in migration? No transformations required?

I'm a bit confused because https://en.wikipedia.org/wiki/Poisson_regression seems to imply a somewhat different interpretation, along the lines of, if "X origin" increases by a factor of e, then...

Joao Santos Silva

Thank you very much,
Regards,
Stan

[Joao Santos Silva]

Dear Stan Peterburgsky,

Your interpretation is valid for small changes in the variables. Transformations are needed for discrete changes.

Best wishes,

Joao

[Stan Peterburgsky]

Thank you very much!
I had a suspicion that, since this is non-linear, the slope of a tangent line is NOT what the Wikipedia article is referring to.