Hi everyone
I am dealing with a panel data with T=9 and N=46. my dependent variables are school enrolments and health indicators- to be estimated separately. my main regressor is remittances per capita with some controls like gdp public expenditure etc. so i want to estimate how remittances per capita influence education and health indicators. I am using system GMM (STATA xtabond2). This is my first time using xtabond2.
1. Is it compulsory to include the lagged dependent variables as a regressor?
2. How will I specify the xtabond2 command with and without the inclusion of lagged dependent variables
3. one of my results has both AR(1) and AR(2) >0.1. Thus they all do not reject the null hypothesis. But i know the AR(1) ideally should be less than 0.05. What do I do?
Thank you.
I am dealing with a panel data with T=9 and N=46. my dependent variables are school enrolments and health indicators- to be estimated separately. my main regressor is remittances per capita with some controls like gdp public expenditure etc. so i want to estimate how remittances per capita influence education and health indicators. I am using system GMM (STATA xtabond2). This is my first time using xtabond2.
1. Is it compulsory to include the lagged dependent variables as a regressor?
2. How will I specify the xtabond2 command with and without the inclusion of lagged dependent variables
3. one of my results has both AR(1) and AR(2) >0.1. Thus they all do not reject the null hypothesis. But i know the AR(1) ideally should be less than 0.05. What do I do?
Thank you.
