[Sebastian Kripfganz]

Try

test _b[d_L,deputyd:_cons] = _b[o_L,officiald:_cons]

[Lin Liu]

what's wrong with this?
. xtdpdgmm L(0/1).slackz munificence complex dynamics area L(0/1).logasset L(0/1).roa agef edu L(0/1).tenure sex agee L(0/1).deputyd ,
> model(fod) collapse gmm(slackz, lag(1 3)) gmm(munificence, lag(0 1)) gmm(complex, lag(0 1)) gmm(dynamics, lag(0 1)) gmm(area, lag(0 1) ) gmm
> (logasset, lag(1 2)) gmm(roa, lag(1 2)) gmm(agef, lag(0 2)) gmm(edu, lag(0 1)) gmm(sex, lag(0 0) ) gmm(tenure, lag(0 3)) gmm(agee, lag(0 0)
> ) gmm(deputyd, lag(1 2)) gmm(area, lag(0 0) model(md)) gmm( sex, lag(0 0) model(md)) gmm(agee, lag(0 0) model(md)) gmm(slackz, lag (1 1) diff
> model(level)) gmm(munificence, lag(0 0) diff model(level)) gmm(complex, lag(0 0) diff model(level)) gmm(dynamics, lag(0 0) diff model(level
> )) gmm(area, lag(0 0) model(level)) gmm( logasset, lag(1 1) diff model(level)) gmm( roa, lag(1 1) diff model(level)) gmm( agef, lag(0 0) diff
> model(level)) gmm(edu, lag(0 0) diff model(level)) gmm(tenure, lag(0 0) diff model(level)) gmm(sex, lag(0 0) model(level)) gmm(agee, lag(0
> 0) model(level)) gmm(deputyd, lag(1 1) diff model(level)) teffects two vce(r) overid auxiliary

Generalized method of moments estimation

Fitting full model:
Step 1 f(b) = .21961324
Step 2 f(b) = .04905417

Fitting reduced model 1:
Step 1 f(b) = .04378765

Fitting reduced model 2:
Step 1 f(b) = .04737229

Fitting reduced model 3:
Step 1 f(b) = .04174331

Fitting reduced model 4:
Step 1 f(b) = .04725824

Fitting reduced model 5:
Step 1 f(b) = .04894016

Fitting reduced model 6:
Step 1 f(b) = .04896063

Fitting reduced model 7:
Step 1 f(b) = .04693545

Fitting reduced model 8:
Step 1 f(b) = .04642574

Fitting reduced model 9:
Step 1 f(b) = .04471508

Fitting reduced model 10:
Step 1 f(b) = .04683552

Fitting reduced model 11:
Step 1 f(b) = .04554952

Fitting reduced model 12:
Step 1 f(b) = .04759769

Fitting reduced model 13:
Step 1 f(b) = .04775263

Fitting reduced model 15:
Step 1 f(b) = .04901515

Fitting reduced model 16:
Step 1 f(b) = .04815328

Fitting reduced model 17:
Step 1 f(b) = .04883611

Fitting reduced model 18:
Step 1 f(b) = .04345726

Fitting reduced model 19:
Step 1 f(b) = .04868351

Fitting reduced model 20:
Step 1 f(b) = .04871824

Fitting reduced model 21:
Step 1 f(b) = .04833418

Fitting reduced model 22:
Step 1 f(b) = .04895118

Fitting reduced model 23:
Step 1 f(b) = .0457555

Fitting reduced model 24:
Step 1 f(b) = .04576668

Fitting reduced model 25:
Step 1 f(b) = .04623669

Fitting reduced model 26:
Step 1 f(b) = .04865459

Fitting reduced model 27:
Step 1 f(b) = .04898542

Fitting reduced model 28:
Step 1 f(b) = .04787909

Fitting reduced model 29:
Step 1 f(b) = .04526437

Fitting reduced model 30:
Step 1 f(b) = .03214391

Fitting no-mdev model:
Step 1 f(b) = .04428694

Fitting no-level model:
Step 1 f(b) = .0033406

Group variable: code Number of obs = 1142
Time variable: year Number of groups = 257

Moment conditions: linear = 49 Obs per group: min = 1
nonlinear = 0 avg = 4.44358
total = 49 max = 8

------------------------------------------------------------------------------
slackz | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
/L.slackz | .712158 .1045299 6.81 0.000 .5072831 .9170328
/munificence | -.3096574 .8710623 -0.36 0.722 -2.016908 1.397593
/complex | .0136566 .0495609 0.28 0.783 -.0834811 .1107942
/dynamics | -5.158069 54.91996 -0.09 0.925 -112.7992 102.4831
/area | -.3154719 .1799491 -1.75 0.080 -.6681656 .0372218
/logasset | 7.155498 2.668933 2.68 0.007 1.924485 12.38651
/L.logasset | -8.373998 2.782808 -3.01 0.003 -13.8282 -2.919795
/roa | -16.33088 8.771663 -1.86 0.063 -33.52302 .8612657
/L.roa | 8.227022 5.280975 1.56 0.119 -2.123498 18.57754
/agef | .1133634 .0963832 1.18 0.240 -.0755442 .3022709
/edu | -.501057 .3031709 -1.65 0.098 -1.095261 .0931471
/tenure | .0153419 .0475422 0.32 0.747 -.077839 .1085228
/L.tenure | .0190134 .0677046 0.28 0.779 -.1136852 .1517119
/sex | .7554232 .468428 1.61 0.107 -.1626789 1.673525
/agee | -.0292742 .0188381 -1.55 0.120 -.0661961 .0076477
/deputyd | 1.565541 1.06155 1.47 0.140 -.515058 3.646141
/L.deputyd | -1.890005 1.075679 -1.76 0.079 -3.998297 .2182872
2005.year | .3553894 .2172804 1.64 0.102 -.0704723 .7812512
2006.year | .304873 .2516098 1.21 0.226 -.1882731 .7980191
2007.year | .2514045 .274264 0.92 0.359 -.2861431 .7889521
2008.year | .6769215 .4010137 1.69 0.091 -.109051 1.462894
2009.year | .747258 .3946966 1.89 0.058 -.0263332 1.520849
2010.year | .9056145 .4358076 2.08 0.038 .0514473 1.759782
2011.year | .7428411 .4927214 1.51 0.132 -.2228752 1.708557
/_cons | 12.84346 5.229984 2.46 0.014 2.592877 23.09404
------------------------------------------------------------------------------
Instruments corresponding to the linear moment conditions:
1, model(fodev):
L1.slackz L2.slackz L3.slackz
2, model(fodev):
munificence L1.munificence
3, model(fodev):
complex L1.complex
4, model(fodev):
dynamics L1.dynamics
5, model(fodev):
area
6, model(fodev):
L1.logasset L2.logasset
7, model(fodev):
L1.roa L2.roa
8, model(fodev):
agef L2.agef
9, model(fodev):
edu L1.edu
10, model(fodev):
sex
11, model(fodev):
tenure L1.tenure L2.tenure L3.tenure
12, model(fodev):
agee
13, model(fodev):
L1.deputyd L2.deputyd
15, model(mdev):
sex
16, model(mdev):
agee
17, model(level):
L1.D.slackz
18, model(level):
D.munificence
19, model(level):
D.complex
20, model(level):
D.dynamics
21, model(level):
area
22, model(level):
L1.D.logasset
23, model(level):
L1.D.roa
24, model(level):
D.agef
25, model(level):
D.edu
26, model(level):
D.tenure
27, model(level):
sex
28, model(level):
agee
29, model(level):
L1.D.deputyd
30, model(level):
2005bn.year 2006.year 2007.year 2008.year 2009.year 2010.year 2011.year
31, model(level):
_cons

. est store d

. xtdpdgmm L(0/1).slackz munificence complex dynamics area L(0/1).logasset L(0/1).roa agef edu tenure sex agee L(0/1).officiald , model
> (fod) collapse gmm(slackz, lag(1 3)) gmm(munificence, lag(0 2)) gmm(complex, lag(0 2)) gmm(dynamics, lag(0 2)) gmm(area, lag(0 2)) gmm(logas
> set, lag(1 3)) gmm(roa, lag(1 3)) gmm(agef, lag(0 2)) gmm(edu, lag(0 1)) gmm( sex, lag(0 2) ) gmm(tenure, lag(0 2)) gmm(agee, lag(0 2)) gmm
> (officiald, lag(1 3)) gmm(area, lag(0 0) model(md)) gmm( sex, lag(0 0) model(md)) gmm(agee, lag(0 0) model(md)) gmm(slackz, lag(1 1) diff mo
> del(level)) gmm(munificence, lag(0 0) diff model(level)) gmm(complex, lag(0 0) diff model(level)) gmm(dynamics, lag(0 0) diff model(level))
> gmm(area, lag(0 0) model(level)) gmm( logasset, lag(1 1) diff model(level)) gmm( roa, lag(1 1) diff model(level)) gmm( agef, lag(0 0) diff mo
> del(level)) gmm(edu, lag(0 0) diff model(level)) gmm(tenure, lag(0 0) diff model(level)) gmm(sex, lag(0 0) model(level)) gmm(agee, lag(0 0)
> model(level)) gmm(officiald, lag(1 1) diff model(level)) teffects two vce(r) overid auxiliary

Generalized method of moments estimation

Fitting full model:
Step 1 f(b) = .21879516
Step 2 f(b) = .0895368

Fitting reduced model 1:
Step 1 f(b) = .07797342

Fitting reduced model 2:
Step 1 f(b) = .07882463

Fitting reduced model 3:
Step 1 f(b) = .07606501

Fitting reduced model 4:
Step 1 f(b) = .07926185

Fitting reduced model 5:
Step 1 f(b) = .08893557

Fitting reduced model 6:
Step 1 f(b) = .08534523

Fitting reduced model 7:
Step 1 f(b) = .08495086

Fitting reduced model 8:
Step 1 f(b) = .08775046

Fitting reduced model 9:
Step 1 f(b) = .07315281

Fitting reduced model 10:
Step 1 f(b) = .07129769

Fitting reduced model 11:
Step 1 f(b) = .08573825

Fitting reduced model 12:
Step 1 f(b) = .07303121

Fitting reduced model 13:
Step 1 f(b) = .08466086

Fitting reduced model 15:
Step 1 f(b) = .08889474

Fitting reduced model 16:
Step 1 f(b) = .08735535

Fitting reduced model 17:
Step 1 f(b) = .08696995

Fitting reduced model 18:
Step 1 f(b) = .08947708

Fitting reduced model 19:
Step 1 f(b) = .08952212

Fitting reduced model 20:
Step 1 f(b) = .08911202

Fitting reduced model 21:
Step 1 f(b) = .08279975

Fitting reduced model 22:
Step 1 f(b) = .08536359

Fitting reduced model 23:
Step 1 f(b) = .08768312

Fitting reduced model 24:
Step 1 f(b) = .08707116

Fitting reduced model 25:
Step 1 f(b) = .08927467

Fitting reduced model 26:
Step 1 f(b) = .08936924

Fitting reduced model 27:
Step 1 f(b) = .08820959

Fitting reduced model 28:
Step 1 f(b) = .08916825

Fitting reduced model 29:
Step 1 f(b) = .08449728

Fitting reduced model 30:
Step 1 f(b) = .06366293

Fitting no-mdev model:
Step 1 f(b) = .08117433

Fitting no-level model:
Step 1 f(b) = .04141188

Group variable: code Number of obs = 1142
Time variable: year Number of groups = 257

Moment conditions: linear = 59 Obs per group: min = 1
nonlinear = 0 avg = 4.44358
total = 59 max = 8

------------------------------------------------------------------------------
slackz | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
/L.slackz | .705772 .1192805 5.92 0.000 .4719866 .9395575
/munificence | -.4403346 .8550542 -0.51 0.607 -2.11621 1.235541
/complex | -.0347037 .0459247 -0.76 0.450 -.1247144 .055307
/dynamics | 59.96618 56.23615 1.07 0.286 -50.25464 170.187
/area | .0132941 .1548448 0.09 0.932 -.2901961 .3167843
/logasset | 4.962371 2.576222 1.93 0.054 -.086932 10.01167
/L.logasset | -5.606148 2.867325 -1.96 0.051 -11.226 .0137069
/roa | -17.11063 5.502293 -3.11 0.002 -27.89493 -6.326336
/L.roa | 10.60277 3.994177 2.65 0.008 2.774331 18.43122
/agef | .0437009 .1054893 0.41 0.679 -.1630544 .2504561
/edu | .0413938 .2862075 0.14 0.885 -.5195625 .6023501
/tenure | .0048285 .0490966 0.10 0.922 -.0913991 .101056
/sex | .3432621 .6145907 0.56 0.576 -.8613135 1.547838
/agee | .0068542 .0191801 0.36 0.721 -.0307382 .0444465
/officiald | 1.023247 .5825883 1.76 0.079 -.1186047 2.165099
/L.officiald | -1.720383 .6982176 -2.46 0.014 -3.088865 -.351902
2005.year | .4571604 .2177951 2.10 0.036 .0302897 .884031
2006.year | .3961391 .2675335 1.48 0.139 -.128217 .9204952
2007.year | .5708944 .254941 2.24 0.025 .0712192 1.07057
2008.year | .7871243 .4278661 1.84 0.066 -.0514777 1.625726
2009.year | .8271088 .4124585 2.01 0.045 .018705 1.635513
2010.year | 1.027484 .4636634 2.22 0.027 .1187209 1.936248
2011.year | .7870365 .5163632 1.52 0.127 -.2250167 1.79909
/_cons | 2.834887 5.29475 0.54 0.592 -7.542632 13.21241
------------------------------------------------------------------------------
Instruments corresponding to the linear moment conditions:
1, model(fodev):
L1.slackz L2.slackz L3.slackz
2, model(fodev):
munificence L1.munificence L2.munificence
3, model(fodev):
complex L1.complex L2.complex
4, model(fodev):
dynamics L1.dynamics L2.dynamics
5, model(fodev):
area L2.area
6, model(fodev):
L1.logasset L2.logasset L3.logasset
7, model(fodev):
L1.roa L2.roa L3.roa
8, model(fodev):
agef L2.agef
9, model(fodev):
edu L1.edu
10, model(fodev):
sex L1.sex L2.sex
11, model(fodev):
tenure L1.tenure L2.tenure
12, model(fodev):
agee L1.agee L2.agee
13, model(fodev):
L1.officiald L2.officiald L3.officiald
15, model(mdev):
sex
16, model(mdev):
agee
17, model(level):
L1.D.slackz
18, model(level):
D.munificence
19, model(level):
D.complex
20, model(level):
D.dynamics
21, model(level):
area
22, model(level):
L1.D.logasset
23, model(level):
L1.D.roa
24, model(level):
D.agef
25, model(level):
D.edu
26, model(level):
D.tenure
27, model(level):
sex
28, model(level):
agee
29, model(level):
L1.D.officiald
30, model(level):
2005bn.year 2006.year 2007.year 2008.year 2009.year 2010.year 2011.year
31, model(level):
_cons

. est store o

. suest d o

Simultaneous results for d, o

Number of obs = 1,142

-------------------------------------------------------------------------------
| Robust
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
--------------+----------------------------------------------------------------
d_L,slackz |
_cons | .712158 .1241785 5.73 0.000 .4687726 .9555434
--------------+----------------------------------------------------------------
d_munificence |
_cons | -.3096574 1.348379 -0.23 0.818 -2.952431 2.333116
--------------+----------------------------------------------------------------
d_complex |
_cons | .0136566 .0717027 0.19 0.849 -.1268782 .1541913
--------------+----------------------------------------------------------------
d_dynamics |
_cons | -5.158069 74.80452 -0.07 0.945 -151.7722 141.4561
--------------+----------------------------------------------------------------
d_area |
_cons | -.3154719 .2452943 -1.29 0.198 -.7962398 .165296
--------------+----------------------------------------------------------------
d_logasset |
_cons | 7.155498 3.376785 2.12 0.034 .5371209 13.77387
--------------+----------------------------------------------------------------
d_L,logasset |
_cons | -8.373998 3.463342 -2.42 0.016 -15.16202 -1.585972
--------------+----------------------------------------------------------------
d_roa |
_cons | -16.33088 9.433864 -1.73 0.083 -34.82091 2.159156
--------------+----------------------------------------------------------------
d_L,roa |
_cons | 8.227022 5.302617 1.55 0.121 -2.165917 18.61996
--------------+----------------------------------------------------------------
d_agef |
_cons | .1133634 .1095392 1.03 0.301 -.1013295 .3280562
--------------+----------------------------------------------------------------
d_edu |
_cons | -.501057 .3479369 -1.44 0.150 -1.183001 .1808867
--------------+----------------------------------------------------------------
d_tenure |
_cons | .0153419 .0756236 0.20 0.839 -.1328777 .1635615
--------------+----------------------------------------------------------------
d_L,tenure |
_cons | .0190134 .0887909 0.21 0.830 -.1550136 .1930403
--------------+----------------------------------------------------------------
d_sex |
_cons | .7554232 .561197 1.35 0.178 -.3445028 1.855349
--------------+----------------------------------------------------------------
d_agee |
_cons | -.0292742 .0263493 -1.11 0.267 -.0809178 .0223694
--------------+----------------------------------------------------------------
d_deputyd |
_cons | 1.565541 2.45065 0.64 0.523 -3.237644 6.368727
--------------+----------------------------------------------------------------
d_L,deputyd |
_cons | -1.890005 2.42004 -0.78 0.435 -6.633196 2.853186
--------------+----------------------------------------------------------------
d_2005,year |
_cons | .3553894 .35439 1.00 0.316 -.3392022 1.049981
--------------+----------------------------------------------------------------
d_2006,year |
_cons | .304873 .3598552 0.85 0.397 -.4004301 1.010176
--------------+----------------------------------------------------------------
d_2007,year |
_cons | .2514045 .4118691 0.61 0.542 -.555844 1.058653
--------------+----------------------------------------------------------------
d_2008,year |
_cons | .6769215 .4867027 1.39 0.164 -.2769983 1.630841
--------------+----------------------------------------------------------------
d_2009,year |
_cons | .747258 .4898263 1.53 0.127 -.212784 1.7073
--------------+----------------------------------------------------------------
d_2010,year |
_cons | .9056145 .5814301 1.56 0.119 -.2339677 2.045197
--------------+----------------------------------------------------------------
d_2011,year |
_cons | .7428411 .5726409 1.30 0.195 -.3795145 1.865197
--------------+----------------------------------------------------------------
d__cons |
_cons | 12.84346 7.006265 1.83 0.067 -.8885694 26.57548
--------------+----------------------------------------------------------------
o_L,slackz |
_cons | .705772 .1116469 6.32 0.000 .4869481 .9245959
--------------+----------------------------------------------------------------
o_munificence |
_cons | -.4403346 1.11817 -0.39 0.694 -2.631908 1.751239
--------------+----------------------------------------------------------------
o_complex |
_cons | -.0347037 .058277 -0.60 0.552 -.1489245 .0795171
--------------+----------------------------------------------------------------
o_dynamics |
_cons | 59.96618 57.7011 1.04 0.299 -53.1259 173.0583
--------------+----------------------------------------------------------------
o_area |
_cons | .0132941 .1780203 0.07 0.940 -.3356192 .3622074
--------------+----------------------------------------------------------------
o_logasset |
_cons | 4.962371 3.533726 1.40 0.160 -1.963606 11.88835
--------------+----------------------------------------------------------------
o_L,logasset |
_cons | -5.606148 3.543803 -1.58 0.114 -12.55187 1.339579
--------------+----------------------------------------------------------------
o_roa |
_cons | -17.11063 6.531034 -2.62 0.009 -29.91122 -4.31004
--------------+----------------------------------------------------------------
o_L,roa |
_cons | 10.60277 4.084201 2.60 0.009 2.597888 18.60766
--------------+----------------------------------------------------------------
o_agef |
_cons | .0437009 .1018201 0.43 0.668 -.1558629 .2432646
--------------+----------------------------------------------------------------
o_edu |
_cons | .0413938 .3063771 0.14 0.893 -.5590943 .6418819
--------------+----------------------------------------------------------------
o_tenure |
_cons | .0048285 .0539864 0.09 0.929 -.100983 .1106399
--------------+----------------------------------------------------------------
o_sex |
_cons | .3432621 .8802884 0.39 0.697 -1.382072 2.068596
--------------+----------------------------------------------------------------
o_agee |
_cons | .0068542 .0200002 0.34 0.732 -.0323456 .0460539
--------------+----------------------------------------------------------------
o_officiald |
_cons | 1.023247 1.430928 0.72 0.475 -1.78132 3.827815
--------------+----------------------------------------------------------------
o_L,officiald |
_cons | -1.720383 1.335567 -1.29 0.198 -4.338046 .8972792
--------------+----------------------------------------------------------------
o_2005,year |
_cons | .4571604 .33954 1.35 0.178 -.2083257 1.122646
--------------+----------------------------------------------------------------
o_2006,year |
_cons | .3961391 .3428259 1.16 0.248 -.2757874 1.068066
--------------+----------------------------------------------------------------
o_2007,year |
_cons | .5708944 .3403995 1.68 0.094 -.0962763 1.238065
--------------+----------------------------------------------------------------
o_2008,year |
_cons | .7871243 .4595304 1.71 0.087 -.1135387 1.687787
--------------+----------------------------------------------------------------
o_2009,year |
_cons | .8271088 .476522 1.74 0.083 -.1068572 1.761075
--------------+----------------------------------------------------------------
o_2010,year |
_cons | 1.027484 .5088123 2.02 0.043 .0302307 2.024738
--------------+----------------------------------------------------------------
o_2011,year |
_cons | .7870365 .5419654 1.45 0.146 -.2751962 1.849269
--------------+----------------------------------------------------------------
o__cons |
_cons | 2.834887 6.433989 0.44 0.659 -9.7755 15.44527
-------------------------------------------------------------------------------

. test [d_L,deputyd=o_L,officiald]
last test not found
r(302);

. test d_L,deputyd=o_L,officiald
invalid 'officiald'
r(198);

. test d_L.deputyd=o_L.officiald
d_L: operator invalid
r(198);

. test [d_L.deputyd=o_L.officiald]
equation d_L not found
r(303);

.

[Sebastian Kripfganz]

Assuming that the coefficients are at a comparable scale, you could probably use the suest command to combine the two regressions. Then you can test for equality of the coefficients with the test command. For this to work, you need to run the two xtdpdgmm regressions with the auxiliary option, store the results with estimates store, and then call suest with the two stored estimation results.

[Lin Liu]

I have two equations of dynamic panel data. One is y=b0+b1x1+b2x2+b3x3, and the other is y=A0+A1x1+A2x4+A3x5. I get the coefficients through the xtdpdgmm. Now, I want to compare b2 and A3 to know which has a larger effect on y. Would you give me some advice?

[Sebastian Kripfganz]

xtdpdgmm does not provide standardized regression coefficients. You would need to standardize all of your variables manually before running the regression.

I can hardly imagine a situation in the context of dynamic panel models where this is meaningful. There is a lot of things that can go wrong and the interpretation of the results becomes anything than straightforward.

[Lin Liu]

Hello Dr. Kripfganz, how can i get standard regression coefficents from your xtdpdgmm? As far as I know，your xtdpdgmm presents nonstandard coefficents.

[Sebastian Kripfganz]

The GMM estimators implemented in xtdpdgmm (or xtabond2) are intended for setting with large N relative to T. Your data set does not really fit into that category. It neither fits into the large-N, large-T world. With small N, you cannot expect to reliably estimate the optimal weighting matrix. Thus, any two-step GMM estimator (that allows for arbitrary correlation within groups over time) is eliminated from the discussion. Similarly, it is not recommended to compute robust standard errors clustered at the group level. You could still use a conventional IV/2SLS estimator with ivregress or xtivreg.

[Piero Best]

Hello Dr. Kripfganz,
I really appreciate all your work.

I am currently having trouble with my dynamic model. N = 15. T = 64.
The empiric evidence in my paper offers a variety of dynamic panels being made and because there is endogeneity I decided to give it a try to your work in order to achieve the best results.

in xtabond2 I had the best model by making all my variables strictly exogenous and only having the lagged depedent variable in the gmm side. But that is not what I intended.

For long panels, which dynamic panel model could I use as an alternative? Or should I just stay with a xtreg fe despite endogeneity being present?

xtabond2 ROA L.ROA creditgrowth Liquidity Provsdefault RatioCapGlobal D.vix Buffer LCR NSFR RR Tier1, gmm(L.ROA. collapse) iv((creditgrowth Liquidity Provsdefault RatioCapGlobal D.vix Buffer LCR NSFR RR Tier1)) robust orthogonal pca

also how can i run this xtabond2 model in xtdpdgmm.

Again i really appreciate all your work, thank you in advance.

[Sebastian Kripfganz]

Maybe the following official video tutorial is of help:
Profile plots and interaction plots in Stata®: Interactions of two continuous variables

[Lin Liu]

I find that your margins and marginsplot can only analyze the interaction of one categorical variable and continuous variable or two categorical variables. But I want to analyze the interaction of two continuous variables, how can I do?

[Sebastian Kripfganz]

I believe you can use margins and marginsplot for that purpose. I have not used them much myself.

[Lin Liu]

Dear,Kripfganz, would you recommend me how to plot the interaction effect of dynamic panel data?

[Sebastian Kripfganz]

I am sorry that the bug had significant effects on your results.

From your results, it seems that there are some issues primarily related to the variables offideputy and offdepdyn. But there is not much more I can say about that. While this is probably not satisfying, sometimes we may just have to live with some imperfection in our models. If you can justify your model specification on theoretical grounds, it might be acceptable to put less emphasis on the specification tests. This also depends on whether the troublesome variables are your main variables of interest or just some control variables. The specification tests then could tell us that we need to be cautious with the interpretation of our results. A perfect model may not exist given the available data.

[Lin Liu]

After you fixed the interaction term bug, the code is the same as the above, but the results are very different.

. xtdpdgmm L(0/1).slackz munificence complex dynamics area L(0/1).logasset L(0/1).roa agef edu tenure sex agee L(0/1).offideputy off
> ideputy#c.dynamics l.offdepdyn, model(fod) collapse gmm(slackz, lag(1 2)) gmm(munificence, lag(0 1)) gmm(complex, lag(0 1)) gmm(dynamics,
> lag(0 1)) gmm(area, lag(0 0)) gmm(logasset, lag(1 2)) gmm(roa, lag(1 2)) gmm(agef, lag(0 1)) gmm(edu, lag(0 0)) gmm(sex, lag(0 0) ) gmm
> (tenure, lag(0 1)) gmm(agee, lag(0 0)) gmm(offideputy, lag(1 2)) gmm(offideputy#c.dynamics, lag(1 2)) gmm(l.offdepdyn, lag(1 2)) gmm(ar
> ea, lag(0 0) model(md)) gmm(edu, lag(0 0) model(md)) gmm( sex, lag(0 0) model(md)) gmm(agee, lag(0 0) model(md)) gmm(slackz, lag(1 1) di
> ff model(level)) gmm(munificence, lag(0 0) diff model(level)) gmm(complex, lag(0 0) diff model(level)) gmm(dynamics, lag(0 0) diff model(
> level)) gmm(area, lag(0 0) model(level)) gmm( logasset, lag(1 1) diff model(level)) gmm( roa, lag(1 1) diff model(level)) gmm( agef, lag(0
> 0) diff model(level)) gmm(edu, lag(0 0) model(level)) gmm(tenure, lag(0 0) diff model(level)) gmm(sex, lag(0 0) model(level)) gmm(ag
> ee , lag(0 0) model(level)) gmm(offideputy, lag(1 1) diff model(level)) gmm(offideputy#c.dynamics, lag(1 1) diff model(level)) gmm(l.offde
> pdyn, lag(1 1) diff model(level)) teffects two vce(r) overid

Generalized method of moments estimation

Fitting full model:
Step 1 f(b) = .1478547
Step 2 f(b) = .08858372

Fitting reduced model 1:
Step 1 f(b) = .07959146

Fitting reduced model 2:
Step 1 f(b) = .08288294

Fitting reduced model 3:
Step 1 f(b) = .06616146

Fitting reduced model 4:
Step 1 f(b) = .07903324

Fitting reduced model 5:
Step 1 f(b) = .08290025

Fitting reduced model 6:
Step 1 f(b) = .08119472

Fitting reduced model 7:
Step 1 f(b) = .08481893

Fitting reduced model 8:
Step 1 f(b) = .08344108

Fitting reduced model 9:
Step 1 f(b) = .08596838

Fitting reduced model 10:
Step 1 f(b) = .08845637

Fitting reduced model 11:
Step 1 f(b) = .08755818

Fitting reduced model 12:
Step 1 f(b) = .08686272

Fitting reduced model 13:
Step 1 f(b) = .0616538

Fitting reduced model 14:
Step 1 f(b) = .05377071

Fitting reduced model 15:
Step 1 f(b) = .07578047

Fitting reduced model 17:
Step 1 f(b) = .08522976

Fitting reduced model 18:
Step 1 f(b) = .08837233

Fitting reduced model 19:
Step 1 f(b) = .08831736

Fitting reduced model 20:
Step 1 f(b) = .08690937

Fitting reduced model 21:
Step 1 f(b) = .08840091

Fitting reduced model 22:
Step 1 f(b) = .08839514

Fitting reduced model 23:
Step 1 f(b) = .08461019

Fitting reduced model 24:
Step 1 f(b) = .07909719

Fitting reduced model 25:
Step 1 f(b) = .08036437

Fitting reduced model 26:
Step 1 f(b) = .08538299

Fitting reduced model 27:
Step 1 f(b) = .08597612

Fitting reduced model 28:
Step 1 f(b) = .08441558

Fitting reduced model 29:
Step 1 f(b) = .08858137

Fitting reduced model 30:
Step 1 f(b) = .08845756

Fitting reduced model 31:
Step 1 f(b) = .08782259

Fitting reduced model 32:
Step 1 f(b) = .08478439

Fitting reduced model 33:
Step 1 f(b) = .08518175

Fitting reduced model 34:
Step 1 f(b) = .08184376

Fitting reduced model 35:
Step 1 f(b) = .03041028

Fitting no-fodev model:
Step 1 f(b) = .00205029

Fitting no-mdev model:
Step 1 f(b) = .06910487

Fitting no-level model:
Step 1 f(b) = .00290704

Group variable: code Number of obs = 1142
Time variable: year Number of groups = 257

Moment conditions: linear = 52 Obs per group: min = 1
nonlinear = 0 avg = 4.44358
total = 52 max = 8

(Std. Err. adjusted for 257 clusters in code)
---------------------------------------------------------------------------------------
| WC-Robust
slackz | Coef. Std. Err. z P>|z| [95% Conf. Interval]
----------------------+----------------------------------------------------------------
slackz |
L1. | .6368078 .1193084 5.34 0.000 .4029676 .870648
|
munificence | -.662565 .78593 -0.84 0.399 -2.202959 .8778295
complex | -.0397233 .051535 -0.77 0.441 -.1407301 .0612835
dynamics | 42.29624 78.94277 0.54 0.592 -112.4287 197.0212
area | -.0145929 .1966068 -0.07 0.941 -.3999352 .3707494
|
logasset |
--. | 6.389995 3.03789 2.10 0.035 .4358389 12.34415
L1. | -7.461377 3.150692 -2.37 0.018 -13.63662 -1.286134
|
roa |
--. | -10.47766 5.867424 -1.79 0.074 -21.9776 1.022282
L1. | 5.698477 3.432466 1.66 0.097 -1.029033 12.42599
|
agef | .0420754 .0928689 0.45 0.651 -.1399443 .2240951
edu | -.0299814 .0794425 -0.38 0.706 -.1856859 .1257231
tenure | .0034255 .0496194 0.07 0.945 -.0938267 .1006777
sex | .3800276 .4998849 0.76 0.447 -.5997288 1.359784
agee | .0020611 .0118086 0.17 0.861 -.0210832 .0252054
|
offideputy |
--. | 1.167884 3.243949 0.36 0.719 -5.190138 7.525906
L1. | -1.071466 4.479092 -0.24 0.811 -9.850325 7.707393
|
offideputy#c.dynamics |
1 | 15.51207 178.1701 0.09 0.931 -333.6949 364.719
|
offdepdyn |
L1. | -94.22112 231.9872 -0.41 0.685 -548.9076 360.4654
|
year |
2005 | .4139393 .2505291 1.65 0.098 -.0770886 .9049672
2006 | .4621536 .2813397 1.64 0.100 -.0892621 1.013569
2007 | .4226051 .3122018 1.35 0.176 -.1892992 1.034509
2008 | .8698897 .4164206 2.09 0.037 .0537203 1.686059
2009 | .6991356 .3981517 1.76 0.079 -.0812274 1.479499
2010 | 1.047527 .4302469 2.43 0.015 .2042586 1.890795
2011 | .8706683 .4820771 1.81 0.071 -.0741855 1.815522
|
_cons | 7.557025 4.05535 1.86 0.062 -.3913147 15.50536
---------------------------------------------------------------------------------------
Instruments corresponding to the linear moment conditions:
1, model(fodev):
L1.slackz L2.slackz
2, model(fodev):
munificence L1.munificence
3, model(fodev):
complex L1.complex
4, model(fodev):
dynamics
5, model(fodev):
area
6, model(fodev):
L1.logasset L2.logasset
7, model(fodev):
L1.roa L2.roa
8, model(fodev):
agef
9, model(fodev):
edu
10, model(fodev):
sex
11, model(fodev):
tenure L1.tenure
12, model(fodev):
agee
13, model(fodev):
L1.offideputy L2.offideputy
14, model(fodev):
L1.(0b.offideputy#c.dynamics) L2.(0b.offideputy#c.dynamics)
L1.(1.offideputy#c.dynamics) L2.(1.offideputy#c.dynamics)
15, model(fodev):
L2.L.offdepdyn
17, model(mdev):
edu
18, model(mdev):
sex
19, model(mdev):
agee
20, model(level):
L1.D.slackz
21, model(level):
D.munificence
22, model(level):
D.complex
23, model(level):
D.dynamics
24, model(level):
area
25, model(level):
L1.D.logasset
26, model(level):
L1.D.roa
27, model(level):
D.agef
28, model(level):
edu
29, model(level):
D.tenure
30, model(level):
sex
31, model(level):
agee
32, model(level):
L1.D.offideputy
33, model(level):
L1.D.(0b.offideputy#c.dynamics) L1.D.(1.offideputy#c.dynamics)
34, model(level):
L1.D.L.offdepdyn
35, model(level):
2005bn.year 2006.year 2007.year 2008.year 2009.year 2010.year 2011.year
36, model(level):
_cons

. estat serial, ar(1/3)

Arellano-Bond test for autocorrelation of the first-differenced residuals
H0: no autocorrelation of order 1: z = -2.6878 Prob > |z| = 0.0072
H0: no autocorrelation of order 2: z = -0.7629 Prob > |z| = 0.4455
H0: no autocorrelation of order 3: z = 0.7649 Prob > |z| = 0.4444

.
. estat overid

Sargan-Hansen test of the overidentifying restrictions
H0: overidentifying restrictions are valid

2-step moment functions, 2-step weighting matrix chi2(26) = 22.7660
Prob > chi2 = 0.6461

2-step moment functions, 3-step weighting matrix chi2(26) = 25.7361
Prob > chi2 = 0.4777

.
. estat overid, difference

Sargan-Hansen (difference) test of the overidentifying restrictions
H0: (additional) overidentifying restrictions are valid

2-step weighting matrix from full model

| Excluding | Difference
Moment conditions | chi2 df p | chi2 df p
------------------+-----------------------------+-----------------------------
1, model(fodev) | 20.4550 24 0.6706 | 2.3110 2 0.3149
2, model(fodev) | 21.3009 24 0.6209 | 1.4651 2 0.4807
3, model(fodev) | 17.0035 24 0.8485 | 5.7625 2 0.0561
4, model(fodev) | 20.3115 25 0.7303 | 2.4545 1 0.1172
5, model(fodev) | 21.3054 25 0.6755 | 1.4607 1 0.2268
6, model(fodev) | 20.8670 24 0.6466 | 1.8990 2 0.3869
7, model(fodev) | 21.7985 24 0.5913 | 0.9676 2 0.6165
8, model(fodev) | 21.4444 25 0.6676 | 1.3217 1 0.2503
9, model(fodev) | 22.0939 25 0.6303 | 0.6721 1 0.4123
10, model(fodev) | 22.7333 25 0.5931 | 0.0327 1 0.8564
11, model(fodev) | 22.5025 24 0.5493 | 0.2636 2 0.8765
12, model(fodev) | 22.3237 25 0.6170 | 0.4423 1 0.5060
13, model(fodev) | 15.8450 24 0.8936 | 6.9210 2 0.0314
14, model(fodev) | 13.8191 22 0.9078 | 8.9469 4 0.0624
15, model(fodev) | 19.4756 25 0.7738 | 3.2904 1 0.0697
17, model(mdev) | 21.9040 25 0.6413 | 0.8620 1 0.3532
18, model(mdev) | 22.7117 25 0.5944 | 0.0543 1 0.8157
19, model(mdev) | 22.6976 25 0.5952 | 0.0685 1 0.7936
20, model(level) | 22.3357 25 0.6163 | 0.4303 1 0.5118
21, model(level) | 22.7190 25 0.5940 | 0.0470 1 0.8284
22, model(level) | 22.7176 25 0.5941 | 0.0485 1 0.8258
23, model(level) | 21.7448 25 0.6504 | 1.0212 1 0.3122
24, model(level) | 20.3280 25 0.7294 | 2.4380 1 0.1184
25, model(level) | 20.6536 25 0.7118 | 2.1124 1 0.1461
26, model(level) | 21.9434 25 0.6390 | 0.8226 1 0.3644
27, model(level) | 22.0959 25 0.6302 | 0.6702 1 0.4130
28, model(level) | 21.6948 25 0.6533 | 1.0712 1 0.3007
29, model(level) | 22.7654 25 0.5913 | 0.0006 1 0.9804
30, model(level) | 22.7336 25 0.5931 | 0.0324 1 0.8571
31, model(level) | 22.5704 25 0.6026 | 0.1956 1 0.6583
32, model(level) | 21.7896 25 0.6479 | 0.9764 1 0.3231
33, model(level) | 21.8917 24 0.5857 | 0.8743 2 0.6459
34, model(level) | 21.0338 25 0.6907 | 1.7322 1 0.1881
35, model(level) | 7.8154 19 0.9884 | 14.9506 7 0.0366
model(fodev) | 0.5269 1 0.4679 | 22.2391 25 0.6219
model(mdev) | 17.7600 23 0.7704 | 5.0061 3 0.1714
model(level) | 0.7471 3 0.8621 | 22.0189 23 0.5191

.
. underid, overid underid kp sw noreport

collinearity check...
collinearities detected in [Y X] (right to left): 0o.offideputy#co.dynamics
collinearities detected in [Y X Z] (right to left): __alliv_51 __alliv_50 __alliv_49 __alliv_48 __alliv_47 __alliv_46 __alliv_45 __alliv_40
> __alliv_39 __alliv_37 __alliv_33 0o.offideputy#co.dynamics
collinearities detected in [X Z Y] (right to left): 2011.year 2010.year 2009.year 2008.year 2007.year 2006.year 2005bn.year 0o.offideputy#co
> .dynamics agee sex edu area
warning: collinearities detected, reparameterization may be advisable

Overidentification test: Kleibergen-Paap robust LIML-based (LM version)
Test statistic robust to heteroskedasticity and clustering on code
j= 19.93 Chi-sq( 25) p-value=0.7506

Underidentification test: Kleibergen-Paap robust LIML-based (LM version)
Test statistic robust to heteroskedasticity and clustering on code
j= 20.84 Chi-sq( 26) p-value=0.7501

2-step GMM J underidentification stats by regressor:
j= 56.19 Chi-sq( 26) p-value=0.0008 L.slackz
j= 111.04 Chi-sq( 26) p-value=0.0000 munificence
j= 85.75 Chi-sq( 26) p-value=0.0000 complex
j= 23.28 Chi-sq( 26) p-value=0.6700 dynamics
j= 24.35 Chi-sq( 26) p-value=0.6107 area
j= 28.99 Chi-sq( 26) p-value=0.3614 logasset
j= 27.10 Chi-sq( 26) p-value=0.4585 L.logasset
j= 39.83 Chi-sq( 26) p-value=0.0532 roa
j= 36.92 Chi-sq( 26) p-value=0.0965 L.roa
j= 44.47 Chi-sq( 26) p-value=0.0185 agef
j= 41.20 Chi-sq( 26) p-value=0.0394 edu
j= 36.91 Chi-sq( 26) p-value=0.0967 tenure
j= 24.53 Chi-sq( 26) p-value=0.6007 sex
j= 31.68 Chi-sq( 26) p-value=0.2440 agee
j= 11.02 Chi-sq( 26) p-value=0.9972 offideputy
j= 19.34 Chi-sq( 26) p-value=0.8574 L.offideputy
j= 11.02 Chi-sq( 26) p-value=0.9972 0b.offideputy#co.dynamics
j= 11.02 Chi-sq( 26) p-value=0.9972 1.offideputy#c.dynamics
j= 17.37 Chi-sq( 26) p-value=0.9217 L.offdepdyn
j= 74.48 Chi-sq( 26) p-value=0.0000 2005bn.year
j= 74.48 Chi-sq( 26) p-value=0.0000 2006.year
j= 74.48 Chi-sq( 26) p-value=0.0000 2007.year
j= 74.48 Chi-sq( 26) p-value=0.0000 2008.year
j= 74.48 Chi-sq( 26) p-value=0.0000 2009.year
j= 74.48 Chi-sq( 26) p-value=0.0000 2010.year
j= 74.48 Chi-sq( 26) p-value=0.0000 2011.year

[Sebastian Kripfganz]

From a programmer's perspective, factor variables and interaction terms are some of the nastiest animals in the Stata universe. They always find a way to escape your carefully designed algorithms. So it happened again with xtdpdgmm. There was unfortunately an annoying bug that could result in incorrect estimates when interaction terms were specified as instruments. I hopefully now managed to tame these animals once and for all with the latest bug fix.

Version 2.3.1 is now available on my personal website and on SSC (with the usual thanks to Kit Baum).

adoupdate xtdpdgmm, update