The RATS Software Forum

Hi Tom,

I ran this GMM :

@ablags(collapse) policy abpolicy
instruments abpolicy dgrowth{2 3 4} dinaxindep1{1 2 3} dinaxfsi{2 3 4} dinaxopen1{1 2 4} dinaxfiscal3{1 2 3}
*
inquire(reglist,valid=absample)
#dpolicy{2} dgrowth{2 3 4} dinaxfsi{2 3 4} dinaxopen1{1 2 3} dinaxindep1{1 2 3} dinaxfiscal3{1 2 3}
mcov(instruments,lwform=||-1.0,2.0,-1.0||,smpl=absample)
linreg(inst,wmatrix=inv(%cmom),robusterrors,lwindow=panel,$
title="Arellano-Bond One-Step Estimator") policy
#policy{1} growth{1} inaxfsi11 inaxindep1 inaxfiscal3 inaxopen1

ad got this result:

Linear Regression - Estimation by Arellano-Bond One-Step Estimator
With Clustered Standard Error Calculations
Dependent Variable POLICY
Panel(64) of Quarterly Data From 1//1995:01 To 17//2009:04
Usable Observations 847
Degrees of Freedom 841
Skipped/Missing (from 1084) 237
Mean of Dependent Variable 6.4300826446
Std Error of Dependent Variable 5.5887466183
Standard Error of Estimate 1.7179282662
Sum of Squared Residuals 2482.0244008
J-Specification(72) 15.0000
Significance Level of J 1.0000000
Durbin-Watson Statistic 2.2188

Variable Coeff Std Error T-Stat Signif
************************************************************************************
1. POLICY{1} 0.923736969 0.041418843 22.30234 0.00000000
2. GROWTH{1} -0.004858936 0.028863025 -0.16834 0.86631214
3. INAXFSI11 -0.012578940 0.005864544 -2.14491 0.03195977
4. INAXINDEP1 0.082452108 0.114935138 0.71738 0.47313999
5. INAXFISCAL3 -3.064695178 1.228778249 -2.49410 0.01262772
6. INAXOPEN1 0.336338456 0.084003334 4.00387 0.00006231


No matter how I vary the instruments, I alwasy get a J-Specification of 15.00000 and a p-value of 1.000

What could be wrong???????????

I's appreciate any guidance on this

regards
dnfloro1

You would need to post the program and data set. That doesn't happen with other data.