Initial value of the dependent variable relation type

[Omar Sattar]

Hello everyone,

I'm currently assessing the impact of financial openness on growth with a panel regression.
My dependent variable is GDP and I use the initial value of the GDP (INGDP) at the start year of the study for each country among the set of the explanatory variables in order to control for the convergence effect.
The question is, the relation between INGDP and GDP should be negative but i got a positive relation in my result ?
I've tried different approaches with my data but i still receive a positive relation.

Any insights ?

Thanks

[Gabriel Temesgen]

Dear @Omar Sattar, could you guide me how to include initial value of the depednet variable [initial value of the GDP (INGDP)] as a regressor. I am also planning to use in my study but failed to understand how to include it.
Thanks