Dear Professors and Researchers,

I’m working on panel data for 156 countries during the period (1960 – 2014).

I have to estimate the following two equations:

  • Prb(Ethnic inclusion) = B1 Oil Rents + B2 GDP Per Capita + B3 Urbanization
  • GDP Per Capita = B1 Ethnic inclusion + B2 Oil Rents + B3 Urbanization

As you can see, (Ethnic Inclusion) and (GDP Per Capita) appear once as independent and once as dependent variable in the two equations. Therefore, these two variables are endogenous & I believe that I need to use Simultaneous Equation Model to estimate the parameters.

These are the steps that I was thinking to follow in order to estimate the model in Stata. I need your advice in order to correct the steps or to tell me another way to estimate these two equations.

1- Estimate the 1st equation using xtprobit (as the dependent variable is a dummy variable).

2- Predict the 1st equation using the code: predict EthnicInclusion_predict e(sample)

3- Replace the 1st variable of the second equation (which is ethnic inclusion) with the predicted value of the 1st equation.

Is this a solution to solve the simultaneity bias? If not, how can I estimate these two equations in Stata? Thank you so much in advance. I will appreciate any advice as it’s my 1st time to work with Simultaneous Equation Models.

Sincerely,

More Dina Abdelsalam's questions See All
Similar questions and discussions