GMM for a system of equations (TSP --> EViews 9))
Posted: Mon Jan 25, 2016 11:38 am
Hello,
I need to estimate a system of equations with GMM for panel data in Eviews9: y(it) = c_1(i) + c_2(i)*x_2 +c_3*x_3(i) @ c inst_1 inst_2 inst_3, where I resrict c_3 to be the same for all i (Wald test does not reject this constraint).
Now, my questions are:
1) What would be the appropriate panel GMM options closest to the following specification written in TSP, i.e., I need the translation of the TSP code into EViews?
TSP code: gmm(nohetero, ..., nma=2), where the stated options mean the following:
HETERO/NOHETERO specifies conditional heteroskedasticity of the residuals, and causes the COVOC matrix to include interaction terms of the residuals and the derivatives with respect to the parameters.
Specify this option to obtain the usual Hansen or Chamberlain estimator. When HETERO is on, GMM checks that the number of OC's is less that the number of observations so that COVOC will be positive definite
(if not, an error message is printed).
NMA= number of autocorrelation terms (AR and/or MA) to be used in computing COVOC. Some forecasting-type models imply a given NMA value, but other models have no natural choice.
See the Andrews reference for automatic bandwidth selection procedures. When there are missing values in the series, NMA does not include terms which cross the gaps in
the data. This is useful in panel data estimation.
2) If Eviews includes the constant term to the list of instruments automatically, is it a c_i term analog to the set of cross-section dummies?
3) How can I test, whether I correctly specified the GMM and GLS weighting, as well as the coefficient covariance method?
I kindly thank you for your nice consideration!
I need to estimate a system of equations with GMM for panel data in Eviews9: y(it) = c_1(i) + c_2(i)*x_2 +c_3*x_3(i) @ c inst_1 inst_2 inst_3, where I resrict c_3 to be the same for all i (Wald test does not reject this constraint).
Now, my questions are:
1) What would be the appropriate panel GMM options closest to the following specification written in TSP, i.e., I need the translation of the TSP code into EViews?
TSP code: gmm(nohetero, ..., nma=2), where the stated options mean the following:
HETERO/NOHETERO specifies conditional heteroskedasticity of the residuals, and causes the COVOC matrix to include interaction terms of the residuals and the derivatives with respect to the parameters.
Specify this option to obtain the usual Hansen or Chamberlain estimator. When HETERO is on, GMM checks that the number of OC's is less that the number of observations so that COVOC will be positive definite
(if not, an error message is printed).
NMA= number of autocorrelation terms (AR and/or MA) to be used in computing COVOC. Some forecasting-type models imply a given NMA value, but other models have no natural choice.
See the Andrews reference for automatic bandwidth selection procedures. When there are missing values in the series, NMA does not include terms which cross the gaps in
the data. This is useful in panel data estimation.
2) If Eviews includes the constant term to the list of instruments automatically, is it a c_i term analog to the set of cross-section dummies?
3) How can I test, whether I correctly specified the GMM and GLS weighting, as well as the coefficient covariance method?
I kindly thank you for your nice consideration!