Hi,
I want to calculate Sargan's criteria (S) to choose between two alternative equations (1 and 2).
S=(eu/ g eu)t
Where g is the geometric mean of the dependent variable of equation 2.
If S<1 we choose equation 2 as better model (Say, linear-linear)
If S>1 we choose equation 1 as better model (Say, log-linear)
But I'm not able to trace the correct code and info in the state!
I know there is one Sargan statistics for overidentifying restriction, not sure what to do?
Please advise,
I want to calculate Sargan's criteria (S) to choose between two alternative equations (1 and 2).
S=(eu/ g eu)t
Where g is the geometric mean of the dependent variable of equation 2.
If S<1 we choose equation 2 as better model (Say, linear-linear)
If S>1 we choose equation 1 as better model (Say, log-linear)
But I'm not able to trace the correct code and info in the state!
I know there is one Sargan statistics for overidentifying restriction, not sure what to do?
Please advise,
