Choosing the Appropriate Regression for a Dependent Variable with Zeros (90%) and Positive Values(10%)

[KH FAF]

Hello everyone,
I am currently conducting panel data analysis using Stata and have encountered a specific situation regarding my dependent variable. In the attached file, you will find the description of the variable.
The uniqueness lies in the fact that this variable has observations with both many zero values and few positive values. After exploring various options, I am wondering if the xttobit regression is the most suitable for this type of variable. Here is how I am considering using it:
xttobit y x1 x2 x3 x4 x5 i.year i.country i.sector
And the options for xttobit: vce(oim), vce(bootstrap), and vce(jackknife).
However, I am unsure if this is the best approach in this case. Does anyone have recommendations on the most appropriate regression for this type of mixed variable (zeros and positive values)? Or is there another method I should consider?

I would be grateful for any advice or suggestions. Thank you in advance for your assistance.

[KH FAF]

I chose Tobit regression due to the exclusive presence of positive values in my dependent variable. Specifically, Y is assigned the value of Z (residuals) only if both Z and B are greater than zero; and zero otherwise. Consequently, I think that my variable is left-censoring.