Export multiple xtabond2 results

[sad Abu alim]

Hello Statalist users,

I need help with exporting multiple xtbond2 estimations along with their post estimations results (AR1, AR2, Sargan, Hansen) into Word. publication quality level.
example of my multiple commands

xtabond2 L(0/1).ZS3 c.OBSR##c.EA_w Size AssetG ROA LoanAsset LiqR DepFL l.GDPG l.T10Y3M y*,gmm(L.ZS3 OBSR LiqR,laglimits(1 1)) iv(l.GDPG l.T10Y3M y*,eq(level)) iv(l.GDPG l.T10Y3M,eq(diff)) twostep robust small

xtabond2 L(0/1).ZS3 c.OBSR##c.EA_w Size AssetG ROA LoanAsset LiqR LLPLR DepFL l.GDPG l.T10Y3M y*,gmm(L.ZS3 OBSR LiqR,laglimits(1 1)) iv(l.GDPG l.T10Y3M y*,eq(level)) iv(l.GDPG l.T10Y3M,eq(diff)) twostep robust small

xtabond2 L(0/1).ZS3 c.OBSR##c.EA_w Size AssetG ROA LoanAsset LiqR LLPAR DepFL l.GDPG l.T10Y3M y*,gmm(L.ZS3 OBSR LiqR,laglimits(1 1)) iv(l.GDPG l.T10Y3M y*,eq(level)) iv(l.GDPG l.T10Y3M,eq(diff)) twostep robust small

and here is sample of the results i get with each regression

xtabond2 L(0/1).ZS3 c.OBSR##c.EA_w Size AssetG  ROA LoanAsset LiqR DepFL  l.GDPG l.T10Y3M y*,gmm(L.ZS3 OBSR LiqR,laglimits(1 1)) iv(l.GDPG  l.T10Y3M y*,eq(level)) iv(l.GDPG l.T10Y3M,eq(diff)) twostep robust small
Favoring space over speed. To switch, type or click on mata: mata set matafavor speed, perm.
Warning: Two-step estimated covariance matrix of moments is singular.
  Using a generalized inverse to calculate optimal weighting matrix for two-step estimation.
  Difference-in-Sargan/Hansen statistics may be negative.

Dynamic panel-data estimation, two-step system GMM
------------------------------------------------------------------------------
Group variable: id                              Number of obs      =      2766
Time variable : Year                            Number of groups   =       336
Number of instruments = 61                      Obs per group: min =         1
F(22, 335)    =  11490.11                                      avg =      8.23
Prob > F      =     0.000                                      max =         9
-------------------------------------------------------------------------------
              |              Corrected
          ZS3 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
          ZS3 |
          L1. |   .9606215   .0519257    18.50   0.000     .8584799    1.062763
              |
         OBSR |   .0166858   .0074719     2.23   0.026     .0019881    .0313835
         EA_w |   .0234135   .0110084     2.13   0.034     .0017592    .0450677
              |
c.OBSR#c.EA_w |  -.0015945   .0006956    -2.29   0.023    -.0029629   -.0002261
              |
         Size |   .0499903   .0244027     2.05   0.041     .0019885    .0979921
       AssetG |  -.0023591   .0018319    -1.29   0.199    -.0059625    .0012443
          ROA |   .0040595   .0352507     0.12   0.908     -.065281    .0734001
    LoanAsset |   .0031209   .0038364     0.81   0.417    -.0044255    .0106672
         LiqR |  -.0025142   .0032188    -0.78   0.435    -.0088458    .0038174
        DepFL |   .0066999   .0037298     1.80   0.073    -.0006368    .0140366
              |
         GDPG |
          L1. |  -.4352625   .2448995    -1.78   0.076     -.916997    .0464721
              |
       T10Y3M |
          L1. |   -.019525   .0240011    -0.81   0.417    -.0667369     .027687
              |
       yearD1 |          0  (omitted)
       yearD2 |          0  (omitted)
       yearD3 |  -.4798001   .2566407    -1.87   0.062    -.9846305    .0250302
       yearD4 |  -.2549777   .1369852    -1.86   0.064    -.5244372    .0144819
       yearD5 |  -.4169946   .2253129    -1.85   0.065     -.860201    .0262117
       yearD6 |  -.1418621    .053773    -2.64   0.009    -.2476374   -.0360869
       yearD7 |   .0177436   .0359721     0.49   0.622     -.053016    .0885032
       yearD8 |  -.5419371   .3025044    -1.79   0.074    -1.136985    .0531103
       yearD9 |  -.1840684   .1093741    -1.68   0.093    -.3992151    .0310783
      yearD10 |          0  (omitted)
        _cons |          0  (omitted)
-------------------------------------------------------------------------------
Instruments for first differences equation
  Standard
    D.(L.GDPG L.T10Y3M)
  GMM-type (missing=0, separate instruments for each period unless collapsed)
    L.(L.ZS3 OBSR LiqR)
Instruments for levels equation
  Standard
    L.GDPG L.T10Y3M yearD1 yearD2 yearD3 yearD4 yearD5 yearD6 yearD7 yearD8
    yearD9 yearD10
    _cons
  GMM-type (missing=0, separate instruments for each period unless collapsed)
    D.(L.ZS3 OBSR LiqR)
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z =  -5.60  Pr > z =  0.000
Arellano-Bond test for AR(2) in first differences: z =   0.19  Pr > z =  0.848
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(38)   = 210.17  Prob > chi2 =  0.000
  (Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(38)   =  44.19  Prob > chi2 =  0.227
  (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:
  GMM instruments for levels
    Hansen test excluding group:     chi2(12)   =  16.51  Prob > chi2 =  0.169
    Difference (null H = exogenous): chi2(26)   =  27.68  Prob > chi2 =  0.374
  iv(L.GDPG L.T10Y3M yearD1 yearD2 yearD3 yearD4 yearD5 yearD6 yearD7 yearD8 yearD9 yearD10, eq(level))
    Hansen test excluding group:     chi2(29)   =  38.98  Prob > chi2 =  0.102
    Difference (null H = exogenous): chi2(9)    =   5.20  Prob > chi2 =  0.816
  iv(L.GDPG L.T10Y3M, eq(diff))
    Hansen test excluding group:     chi2(36)   =  42.25  Prob > chi2 =  0.219
    Difference (null H = exogenous): chi2(2)    =   1.94  Prob > chi2 =  0.380

what is the best command to do so?

Thank you,
Sad

[Arthur Morris]

It’s always helpful to share what you’ve tried and where you’ve hit a snag.

The best command for this sort of thing is a matter of preference, but I tend to use esttab

Since this is a user-generated command, start with:

ssc install esttab
help esttab

And the extensive web documentation.

I’m not sure about the integration with word, but I’ve found the integration with LaTeX to be satisfactory.

[sad Abu alim]

Thank you Arthur for the timely respond. I will try your advice using esttab. does it work with excel ?
I did try

outreg2 using myfile.doc, se bdec(3) e(ar1p ar2p hansenp sarganp) word replace

it works ok for single regression results. I had issues for multiple estimations.

Best,
Sad

[arshad hayat]

sad Abu alim for multiple equations, you simply replace the word "replace" with append in the 2nd, 3rd.....equation

[Attaullah Shah]

You can also use asdoc, which can be used with almost all Stata commands. For nested tables, just add nest after comma. The orange colors in the following code show addition of asdoc to generate the following table. Here is a YouTube video on changing different options of asdoc when creating regression tables.https://www.youtube.com/watch?v=61ks3cMPz3c

* Install asdoc
ssc install asdoc, replace
use http://www.stata-press.com/data/r7/abdata.dta, clear
asdoc xtabond2 n l.n l(0/1).(w k) yr1980-yr1984, gmm(l.n w k) iv(yr1980-yr1984, passthru) noleveleq small nest replace
asdoc xtabond2 n l.n l(0/1).(w k) yr1980-yr1984, gmm(l.n w k) iv(yr1980-yr1984, mz) robust twostep small h(2) nest
asdoc xtabond2 n l(1/2).n l(0/1).w l(0/2).(k ys) yr1980-yr1984, gmm(l.n w k) iv(yr1980-yr1984) robust twostep small nest

[sad Abu alim]

sad Abu alim for multiple equations, you simply replace the word "replace" with append in the 2nd, 3rd.....equation

arshad,
Thank you for your reply i will try this.

Best wishes,
Sad

[sad Abu alim]

You can also use asdoc, which can be used with almost all Stata commands. For nested tables, just add nest after comma. The orange colors in the following code show addition of asdoc to generate the following table. Here is a YouTube video on changing different options of asdoc when creating regression tables.https://www.youtube.com/watch?v=61ks3cMPz3c

* Install asdoc
ssc install asdoc, replace
use http://www.stata-press.com/data/r7/abdata.dta, clear
asdoc xtabond2 n l.n l(0/1).(w k) yr1980-yr1984, gmm(l.n w k) iv(yr1980-yr1984, passthru) noleveleq small nest replace
asdoc xtabond2 n l.n l(0/1).(w k) yr1980-yr1984, gmm(l.n w k) iv(yr1980-yr1984, mz) robust twostep small h(2) nest
asdoc xtabond2 n l(1/2).n l(0/1).w l(0/2).(k ys) yr1980-yr1984, gmm(l.n w k) iv(yr1980-yr1984) robust twostep small nest

[ATTACH=CONFIG]n1569005[/ATTACH]

Dr Attallah,
Thank you for Asdoc i use it for OLS FE and RE estimations and for Descriptive statistics tables it works great. i will try it for the xtabond2 now after your clarification.

kind regards,
Sad

[Beri Parfait]

I know this comment is late but I came across this formula that makes this really easy as I was struggling to export results. It follows three simple stages:

1. Run your dynamic GMM models y= f(y_1, x), starting with eststo: It could be just one model or many. Check the following example

eststo: xtabond2 FDI FDI GDP  TRADE  RENTS EXR  yr*, gmm(l.FDI  l(0/1).BIT l(0/1).GDP l(0/1).TRADE  l(0/2).RENTS l(0/1).EXR, collapse lag(1 1)) iv(BIT GDP TRADE  RENTS EXR  yr*, eq(level)) twostep robust small orthog

You may write down your codes for multiple models and run them at the same time following the normal ways. In step 2, simply run the following code to display the results on your screen in tabular form.

2. esttab

Take note that at this stage, post estimation results would not be displayed on the table.

3. The third stage depends on where you wish to export your results. It could be Latex (tex), word (rtf), excel or any other. Run the following code

esttab using results.rtf ,label se starlevels( * 0.10 ** 0.05 *** 0.010)  stats(N j ar1p ar2p hansenp, labels("Observations" "No. of instruments" "AR1 (p-value)" "AR2 (p-value)" "Hansen-J (p-value)" "F Statistic"))

If you are using recent versions of Stata, you can open the results directly from your results window, otherwise, go to the folder where the results were sent or where you saved your data. Take note of the dot(.) between results and rtf (rich text format) in the esttab command. Recall that esttab should be installed from

 ssc install esttab

Should you not be interested in exporting results at 10 per cent level of significance, adjust the formula by removing it.

I hope this helps someone in future

[sad Abu alim]

Thank you beri for your contribution to my question.