Hi all,

I want to use Abrigo and Love's pvar package to estimate some panel VAR. My understanding is that pvar calls a GMM estimator without a constant because variables are transformed such that their means are zero. However, if I subtract a constant from variables, the results seem to change which does not look right. Am I missing something or doing something wrong? I use Stata 14.2 for Windows.

To be specific, consider the example used in Section 4.1 of the article on the Stata Journal:

Code:
. webuse psidextract 
. generate lwks=ln(wks)
. pvar lwks lwage if fem==0, lags(3)


GMM Estimation

Final GMM Criterion Q(b) =  1.11e-32
Initial weight matrix: Identity
GMM weight matrix:     Robust
                                                   No. of obs      =      1584
                                                   No. of panels   =       528
                                                   Ave. no. of T   =     3.000


------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
lwks         |
        lwks |
         L1. |   .0477872   .1816701     0.26   0.793    -.3082796    .4038541
         L2. |  -.1891446   .1002787    -1.89   0.059    -.3856872     .007398
         L3. |  -.0694588   .0554891    -1.25   0.211    -.1782155    .0392979
             |
       lwage |
         L1. |  -.0069066   .0249964    -0.28   0.782    -.0558987    .0420855
         L2. |  -.0206062   .0137029    -1.50   0.133    -.0474633    .0062509
         L3. |  -.0224254   .0141702    -1.58   0.114    -.0501985    .0053476
-------------+----------------------------------------------------------------
lwage        |
        lwks |
         L1. |   .3516101   .2541961     1.38   0.167     -.146605    .8498253
         L2. |   .1322435    .123261     1.07   0.283    -.1093435    .3738306
         L3. |   .0890408    .063914     1.39   0.164    -.0362283    .2143099
             |
       lwage |
         L1. |   .5894378   .0820801     7.18   0.000     .4285638    .7503119
         L2. |   .1818445   .0480188     3.79   0.000     .0877293    .2759597
         L3. |   .1337024   .0367614     3.64   0.000     .0616515    .2057533
------------------------------------------------------------------------------
Instruments : l(1/3).(lwks lwage)
That is the same as what is reported in the publication. However, the results change if I subtract a constant from one of the variables:

Code:
. gen lwks1=lwks-1
. pvar lwks1 lwage if fem==0, lags(3)


Panel vector autoregresssion



GMM Estimation

Final GMM Criterion Q(b) =  5.91e-33
Initial weight matrix: Identity
GMM weight matrix:     Robust
                                                   No. of obs      =      1584
                                                   No. of panels   =       528
                                                   Ave. no. of T   =     3.000


------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
lwks1        |
       lwks1 |
         L1. |   .0879381   .1630894     0.54   0.590    -.2317113    .4075874
         L2. |   -.170683   .0852181    -2.00   0.045    -.3377074   -.0036586
         L3. |  -.0603179    .048435    -1.25   0.213    -.1552487    .0346129
             |
       lwage |
         L1. |  -.0052976   .0255565    -0.21   0.836    -.0553875    .0447922
         L2. |  -.0206299   .0138081    -1.49   0.135    -.0476933    .0064334
         L3. |  -.0226514    .014234    -1.59   0.112    -.0505496    .0052468
-------------+----------------------------------------------------------------
lwage        |
       lwks1 |
         L1. |   .2857115   .2018409     1.42   0.157    -.1098894    .6813123
         L2. |   .1019429   .0975398     1.05   0.296    -.0892316    .2931174
         L3. |   .0740381   .0509953     1.45   0.147    -.0259107     .173987
             |
       lwage |
         L1. |    .586797   .0811002     7.24   0.000     .4278435    .7457505
         L2. |   .1818834   .0483517     3.76   0.000     .0871159     .276651
         L3. |   .1340733   .0368877     3.63   0.000     .0617748    .2063718
------------------------------------------------------------------------------
Instruments : l(1/3).(lwks1 lwage)

Thanks!

Best regards,
Anton