Coefficient test for two different reghdfe regression models (suest not working)

[Jan Arendszoon]

Hello, STATA newbie here.

For my research I need to do a coefficient test between coefficients from two different regression models. Usually this is done with the suest command, however, since I'm using a reghdfe regression, this isn't working, since it cannot predict the score. I saw one forum post where someone had the same issue, but the commands posted there were too complicated for me. I have the following two regression models (excluding the control variables):

Model 1: Reghdfe EarningsPersistenceGAAP GAAPEPS CG FirminGFC FirminCOVID GAAPEPS*GFC GAAPEPS*COVID GAAPEPS*GFC*CG GAAPEPS*COVID*CG, absorb(year industry)

Model 2: Reghdfe EarningsPersistenceNONGAAP NONGAAPEPS CG FirminGFC FirminCOVID NONGAAPEPS*GFC NONGAAPEPS*COVID NONGAAPEPS*GFC*CG NONGAAPEPS*COVID*CG, absorb(year industry)

I would like to test every variable for potential statistical differences.

Does anyone know how I can do a manual workaround (that I can understand and explain if necessary)?

[Clyde Schechter]

The difference between the two models is that where Model1 has GAAP, Model 2 has NONGAAP in a variable name. This leads me to think that what are really trying to do is conduct the regression in two subsets of your population, those characterized as GAAP (whatever that is) and the other being those who are not characterized as GAAP. And every observation is either a GAAP observation or a NONGAAP observation (probably with the variables containing the opposite name having missing values). Am I right?

If my thought is correct, then you have not formed your data set optimally for purpose. Instead of having separate variables for GAAP and NONGAAP, have a single variable that takes on the value 1 in GAAP observations and 0 in the NONGAAP observations. For variables like GAAPEPS and NONGAAPEPS, combine them into a single variable, EPS, which takes on the GAAPEPS value if the observation has GAAP == 1 and the NONGAAPEPS value if the observation has GAAP = 0. Then do a single regression like this using factor variable notation:

reghdfe EarningsPersistence i.GAAP##(x.CG x.FirminGFC x.FirminCOVID x.EPS#x.GFC x.x.EPS#x.COVID x.EPS#x.GFC#x.CT x.EPS#x.COVID#x.CG), ///
     absorb(industry i.GAAP#industry year i.GAAP#year)

Note that, the right hand side variables have all been prefixed with x. You must replace x., in each case, by c. if the variable is continuous, and by i. if the variable is discrete. And it may be more complicated than that: if, for example EPS*COVID is actually an interaction term for variables EPS and FirminCOVID, then you need to properly construct that interaction as x.EPS#x.FirminCOVID instead. Before you proceed, do read -help fvvarlist- so you can understand factor variable notation, how it works, and how to use it.

Anyway, once you have properly constructed the regression using factor variable notation and with the whole thing involving an interaction with i.GAAP, you are home. The coefficients of all of the i.GAAP#whatever terms are the differences between the whatever terms in the GAAP and NONGAAP subset regressions.

[George Ford]

reghdfe and areg won't work with suest.

Clyde offers a clean alternative.

[Jan Arendszoon]

Hi Clyde, thanks for your response.

GAAP and non-GAAP are two different variables. Indeed, I have them both in one dataset, but they are different from each other. Hence, the two different regression models. A litte background on the variables:

GFC and COVID are both dummy variables
The interaction variables are generated by multiplying the independent variable with a dummy variable (and another independent variable if applicable), but that's wrong according to your message?


I came across this stata forum, but I don't understand much from it to be honest: https://www.statalist.org/forums/for...erent-outcomes

I will read -help fvvarlist-, thanks for the suggestion.

[Clyde Schechter]

I understand that GAAP and NONGAAP are two separate variables in your data. I'm questioning whether they should be. From their names it sounds like it should be the case that whenever GAAP == 1, we have NONGAAP == 0, and vice versa. Are you saying that this is not the case? OK, but if it is not the case, why did you name them GAAP and NONGAAP? At best that's confusing.

Anyway, on the assumption that GAAP and NONGAAP are in fact two variables that are not related by each being the opposite of the other, then the approach outlined in #2 will not work. In this case, what you need to do is emulate the -reghdfe- commands in -regress-. You can do that by eliminating the -absorb()- option and adding i.year and i.industry to the list of independent variables.

I still think that you should get rid of your homebrew interaction terms and use factor variable notation instead. So the approach I am recommending looks like this:

//    MODEL 1
regress EarningsPersistenceGAAP c.GAAPEPS##c.GC##i.COVID##i.GFC i.year i.industry
estimates store gaap

//    MODEL 2
regress EarningsPersistenceNONGAAP c.NONGAAPEPS##c.GC##i.COVID##i.GFC i.year i.industry
estimates store nongaap

suest gaap nongaap

In the above I am guessing that the EPS and GC variables are continuous and COVID and GFC are discrete (probably even dichotomous), but change the c.'s and i.'s to whatever is actually correct for these variables. Also, as I am reading your code from #1, it seems to me that what you are actually looking at, if I understand which variables are intended to be interactions of which others, that what you are really trying to model here is a four-way interaction involving (NON)GAAPEPS, GC, COVID, and GFC. I've coded it accordingly here.

Your code in #1 makes no mention of robust or clustered variance estimation. Typically one of those is used in this kind of situation, unless the number of clusters is too small to allow it. Anyway, if you want to use either of those variance estimators, do not apply them to the -regress- commands. Instead, continue to use the default OLS variance estimator in -regress-, but apply the other variance estimation to the -suest- command.

[George Ford]

Are EarningsPersistenceGAAP and EarningsPersistenceNONGAAP measuring different things, or measuring the same thing for different groups (GAAP and NONGAAP). Can a firm be in both groups?

[Jan Arendszoon]

Are EarningsPersistenceGAAP and EarningsPersistenceNONGAAP measuring different things, or measuring the same thing for different groups (GAAP and NONGAAP). Can a firm be in both groups?

Hi George, EarningsPersistence (EP) is measured in the same way for both reporting standards (GAAP and non-GAAP). All the firms in my sample report both GAAP earnings per share (EPS) and non-GAAP earnings per share.

[George Ford]

I still think Clyde's reghdfe approach would work, but you'd need to modify the data a bit. Create a single Y and EPS variable and then have a dummy for GAAP, then estimate all the coefficients with i.GAAP interaction on all the Xs. Or, include the Xs and then have c.GAAP#(c.X) and the interactions measure the difference in coefficients and the t-test is a direct test. You can report the i.GAAP coefficients and then just use the c.GAAP approach to test equality.

This approach seems more straightforward and everything you need comes from one (or two) lines of code, versus a bunch of coefficient tests using suest.

[Jan Arendszoon]

I understand that GAAP and NONGAAP are two separate variables in your data. I'm questioning whether they should be. From their names it sounds like it should be the case that whenever GAAP == 1, we have NONGAAP == 0, and vice versa. Are you saying that this is not the case? OK, but if it is not the case, why did you name them GAAP and NONGAAP? At best that's confusing.

Anyway, on the assumption that GAAP and NONGAAP are in fact two variables that are not related by each being the opposite of the other, then the approach outlined in #2 will not work. In this case, what you need to do is emulate the -reghdfe- commands in -regress-. You can do that by eliminating the -absorb()- option and adding i.year and i.industry to the list of independent variables.

I still think that you should get rid of your homebrew interaction terms and use factor variable notation instead. So the approach I am recommending looks like this:

// MODEL 1
regress EarningsPersistenceGAAP c.GAAPEPS##c.GC##i.COVID##i.GFC i.year i.industry
estimates store gaap

// MODEL 2
regress EarningsPersistenceNONGAAP c.NONGAAPEPS##c.GC##i.COVID##i.GFC i.year i.industry
estimates store nongaap

suest gaap nongaap

In the above I am guessing that the EPS and GC variables are continuous and COVID and GFC are discrete (probably even dichotomous), but change the c.'s and i.'s to whatever is actually correct for these variables. Also, as I am reading your code from #1, it seems to me that what you are actually looking at, if I understand which variables are intended to be interactions of which others, that what you are really trying to model here is a four-way interaction involving (NON)GAAPEPS, GC, COVID, and GFC. I've coded it accordingly here.

Your code in #1 makes no mention of robust or clustered variance estimation. Typically one of those is used in this kind of situation, unless the number of clusters is too small to allow it. Anyway, if you want to use either of those variance estimators, do not apply them to the -regress- commands. Instead, continue to use the default OLS variance estimator in -regress-, but apply the other variance estimation to the -suest- command.

Hi Clyde, thanks for your reply. I can understand where the confusion comes from. I should have explained that better. For my research I need to make a distinction between the Global Financial Crisis and the COVID Crisis which means I can't let all the variables interact with each other. Furthermore, if I understand correctly, I should stop using the reghdfe command and ditch the year fixed effects and industry fixed effects by transforming the reghdfe into a normal reg. Will i.year and i.industry capture the fixed effects in the correct way?

. reg EarningsPersistenceGAAP_w GAAPEPS_w CG GFC COVID c.GAAPEPS##i.GFC c.GAAPEPS##i.COVID c.
> GAAPEPS##i.GFC##c.CG c.GAAPEPS##i.COVID##c.CG  Num_LogTA_w NaturalLogarithmofFA_w Num_LR_w 
> Num_SR_w INF_w INFGrowth_w UR_w URGrowth_w IR_w IRGrowth_w GDP_w GDPGrowth_w i.FiscalYearCC
> M i.SIC_Industry 
note: 1.GFC omitted because of collinearity.
note: GAAPEPS omitted because of collinearity.
note: 1.COVID omitted because of collinearity.
note: GAAPEPS omitted because of collinearity.
note: CG omitted because of collinearity.
note: GAAPEPS omitted because of collinearity.
note: CG omitted because of collinearity.
note: 2009.FiscalYearCCM omitted because of collinearity.
note: 2021.FiscalYearCCM omitted because of collinearity.

      Source |       SS           df       MS      Number of obs   =    15,233
-------------+----------------------------------   F(39, 15193)    =    290.03
       Model |  12554.4278        39  321.908405   Prob > F        =    0.0000
    Residual |  16862.9051    15,193  1.10991279   R-squared       =    0.4268
-------------+----------------------------------   Adj R-squared   =    0.4253
       Total |  29417.3328    15,232  1.93128498   Root MSE        =    1.0535

----------------------------------------------------------------------------------------
EarningsPersistenceG~w | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-----------------------+----------------------------------------------------------------
             GAAPEPS_w |   .6681901   .0116702    57.26   0.000     .6453151    .6910651
                    CG |    .012462   .0063395     1.97   0.049     .0000358    .0248881
                   GFC |  -.0087051   .1684049    -0.05   0.959     -.338799    .3213887
                 COVID |   1.715321   .4143013     4.14   0.000     .9032407    2.527401
               GAAPEPS |   .0206978   .0168235     1.23   0.219    -.0122783    .0536738
                 1.GFC |          0  (omitted)
                       |
         GFC#c.GAAPEPS |
                    1  |  -.2866625   .0623326    -4.60   0.000    -.4088419   -.1644831
                       |
               GAAPEPS |          0  (omitted)
               1.COVID |          0  (omitted)
                       |
       COVID#c.GAAPEPS |
                    1  |  -.0519804   .0227925    -2.28   0.023    -.0966564   -.0073044
                       |
               GAAPEPS |          0  (omitted)
                    CG |          0  (omitted)
                       |
        c.GAAPEPS#c.CG |  -.0072304   .0018884    -3.83   0.000     -.010932   -.0035289
                       |
              GFC#c.CG |
                    1  |  -.0110013    .010655    -1.03   0.302    -.0318863    .0098837
                       |
    GFC#c.GAAPEPS#c.CG |
                    1  |   .0168293   .0054928     3.06   0.002     .0060627    .0275959
                       |
               GAAPEPS |          0  (omitted)
                    CG |          0  (omitted)
                       |
            COVID#c.CG |
                    1  |  -.0255031   .0116383    -2.19   0.028    -.0483156   -.0026907
                       |
  COVID#c.GAAPEPS#c.CG |
                    1  |   .0072397   .0024525     2.95   0.003     .0024325    .0120469
                       |
           Num_LogTA_w |   .0350231   .0088228     3.97   0.000     .0177293    .0523169
NaturalLogarithmofFA_w |  -.0254782   .0132157    -1.93   0.054    -.0513826    .0004263
              Num_LR_w |  -.0245809   .0085321    -2.88   0.004    -.0413049    -.007857
              Num_SR_w |  -.0028228   .0028168    -1.00   0.316    -.0083439    .0026984
                 INF_w |  -.0152702   .0086377    -1.77   0.077    -.0322012    .0016608
           INFGrowth_w |   9.317135   4.603709     2.02   0.043     .2933119    18.34096
                  UR_w |   .0304478   .0243885     1.25   0.212    -.0173567    .0782523
            URGrowth_w |  -.0034263    .087293    -0.04   0.969     -.174531    .1676785
                  IR_w |  -.1400874   .0397134    -3.53   0.000    -.2179304   -.0622444
            IRGrowth_w |   .1033606   .0857643     1.21   0.228    -.0647476    .2714689
                 GDP_w |  -.0001371    .000134    -1.02   0.306    -.0003998    .0001256
           GDPGrowth_w |  -.2230232   1.650276    -0.14   0.893    -3.457763    3.011716
                       |
         FiscalYearCCM |
                 2007  |     .10874   .0450905     2.41   0.016     .0203571    .1971229
                 2008  |   .0594729   .0756395     0.79   0.432    -.0887896    .2077353
                 2009  |          0  (omitted)
                 2010  |   .0831276   .1234739     0.67   0.501     -.158896    .3251513
                 2018  |   1.206829   .2899916     4.16   0.000     .6384109    1.775248
                 2019  |   1.169862   .3201219     3.65   0.000     .5423845    1.797339
                 2020  |  -.4886415     .06704    -7.29   0.000     -.620048    -.357235
                 2021  |          0  (omitted)
                 2022  |   1.851714   .4392351     4.22   0.000     .9907599    2.712667
                       |
          SIC_Industry |
                    2  |   .0953184   .1897156     0.50   0.615    -.2765469    .4671836
                    3  |    .002676   .2224802     0.01   0.990    -.4334119    .4387638
                    4  |   .2374448   .1870519     1.27   0.204    -.1291994    .6040891
                    5  |   .0942036   .1878171     0.50   0.616    -.2739404    .4623476
                    6  |   .2452833    .193118     1.27   0.204    -.1332512    .6238178
                    7  |   .2960358   .1888819     1.57   0.117    -.0741954    .6662671
                    9  |   .1679514   .1878889     0.89   0.371    -.2003334    .5362363
                   10  |   .4052408   .2202281     1.84   0.066    -.0264327    .8369143
                       |
                 _cons |   5.628291   1.372383     4.10   0.000     2.938256    8.318327

This is the result

[Jan Arendszoon]

Thoughts?

 reg EarningsPersistenceGAAP_w GAAPEPS_w CG GFC COVID c.GAAPEPS##i.GFC c.GAAPEPS##i.COVID c.GAAPEPS##i.GFC##c.CG c.GAAPEPS##i.COVID##c.CG Num_LogTA_w NaturalLogarithmofFA_w Num_LR_w Num_SR_w INF_w INFGrowth_w UR_w URGrowth_w IR_w 
> IRGrowth_w GDP_w GDPGrowth_w i.FiscalYearCCM i.SIC_Industry 
note: 1.GFC omitted because of collinearity.
note: GAAPEPS omitted because of collinearity.
note: 1.COVID omitted because of collinearity.
note: GAAPEPS omitted because of collinearity.
note: CG omitted because of collinearity.
note: GAAPEPS omitted because of collinearity.
note: CG omitted because of collinearity.
note: 2009.FiscalYearCCM omitted because of collinearity.
note: 2021.FiscalYearCCM omitted because of collinearity.

      Source |       SS           df       MS      Number of obs   =    15,233
-------------+----------------------------------   F(39, 15193)    =    290.03
       Model |  12554.4278        39  321.908405   Prob > F        =    0.0000
    Residual |  16862.9051    15,193  1.10991279   R-squared       =    0.4268
-------------+----------------------------------   Adj R-squared   =    0.4253
       Total |  29417.3328    15,232  1.93128498   Root MSE        =    1.0535

----------------------------------------------------------------------------------------
EarningsPersistenceG~w | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-----------------------+----------------------------------------------------------------
             GAAPEPS_w |   .6681901   .0116702    57.26   0.000     .6453151    .6910651
                    CG |    .012462   .0063395     1.97   0.049     .0000358    .0248881
                   GFC |  -.0087051   .1684049    -0.05   0.959     -.338799    .3213887
                 COVID |   1.715321   .4143013     4.14   0.000     .9032407    2.527401
               GAAPEPS |   .0206978   .0168235     1.23   0.219    -.0122783    .0536738
                 1.GFC |          0  (omitted)
                       |
         GFC#c.GAAPEPS |
                    1  |  -.2866625   .0623326    -4.60   0.000    -.4088419   -.1644831
                       |
               GAAPEPS |          0  (omitted)
               1.COVID |          0  (omitted)
                       |
       COVID#c.GAAPEPS |
                    1  |  -.0519804   .0227925    -2.28   0.023    -.0966564   -.0073044
                       |
               GAAPEPS |          0  (omitted)
                    CG |          0  (omitted)
                       |
        c.GAAPEPS#c.CG |  -.0072304   .0018884    -3.83   0.000     -.010932   -.0035289
                       |
              GFC#c.CG |
                    1  |  -.0110013    .010655    -1.03   0.302    -.0318863    .0098837
                       |
    GFC#c.GAAPEPS#c.CG |
                    1  |   .0168293   .0054928     3.06   0.002     .0060627    .0275959
                       |
               GAAPEPS |          0  (omitted)
                    CG |          0  (omitted)
                       |
            COVID#c.CG |
                    1  |  -.0255031   .0116383    -2.19   0.028    -.0483156   -.0026907
                       |
  COVID#c.GAAPEPS#c.CG |
                    1  |   .0072397   .0024525     2.95   0.003     .0024325    .0120469
                       |
           Num_LogTA_w |   .0350231   .0088228     3.97   0.000     .0177293    .0523169
NaturalLogarithmofFA_w |  -.0254782   .0132157    -1.93   0.054    -.0513826    .0004263
              Num_LR_w |  -.0245809   .0085321    -2.88   0.004    -.0413049    -.007857
              Num_SR_w |  -.0028228   .0028168    -1.00   0.316    -.0083439    .0026984
                 INF_w |  -.0152702   .0086377    -1.77   0.077    -.0322012    .0016608
           INFGrowth_w |   9.317135   4.603709     2.02   0.043     .2933119    18.34096
                  UR_w |   .0304478   .0243885     1.25   0.212    -.0173567    .0782523
            URGrowth_w |  -.0034263    .087293    -0.04   0.969     -.174531    .1676785
                  IR_w |  -.1400874   .0397134    -3.53   0.000    -.2179304   -.0622444
            IRGrowth_w |   .1033606   .0857643     1.21   0.228    -.0647476    .2714689
                 GDP_w |  -.0001371    .000134    -1.02   0.306    -.0003998    .0001256
           GDPGrowth_w |  -.2230232   1.650276    -0.14   0.893    -3.457763    3.011716
                       |
         FiscalYearCCM |
                 2007  |     .10874   .0450905     2.41   0.016     .0203571    .1971229
                 2008  |   .0594729   .0756395     0.79   0.432    -.0887896    .2077353
                 2009  |          0  (omitted)
                 2010  |   .0831276   .1234739     0.67   0.501     -.158896    .3251513
                 2018  |   1.206829   .2899916     4.16   0.000     .6384109    1.775248
                 2019  |   1.169862   .3201219     3.65   0.000     .5423845    1.797339
                 2020  |  -.4886415     .06704    -7.29   0.000     -.620048    -.357235
                 2021  |          0  (omitted)
                 2022  |   1.851714   .4392351     4.22   0.000     .9907599    2.712667
                       |
          SIC_Industry |
                    2  |   .0953184   .1897156     0.50   0.615    -.2765469    .4671836
                    3  |    .002676   .2224802     0.01   0.990    -.4334119    .4387638
                    4  |   .2374448   .1870519     1.27   0.204    -.1291994    .6040891
                    5  |   .0942036   .1878171     0.50   0.616    -.2739404    .4623476
                    6  |   .2452833    .193118     1.27   0.204    -.1332512    .6238178
                    7  |   .2960358   .1888819     1.57   0.117    -.0741954    .6662671
                    9  |   .1679514   .1878889     0.89   0.371    -.2003334    .5362363
                   10  |   .4052408   .2202281     1.84   0.066    -.0264327    .8369143
                       |
                 _cons |   5.628291   1.372383     4.10   0.000     2.938256    8.318327
----------------------------------------------------------------------------------------

. 
. eststo gaap

. 
. 
. 
. reg EarningsPersistenceNONGAAP_w NONGAAPEPS_w CG GFC COVID c.NONGAAPEPS##i.GFC c.NONGAAPEPS##i.COVID c.NONGAAPEPS##i.GFC##c.CG c.NONGAAPEPS##i.COVID##c.CG Num_LogTA_w NaturalLogarithmofFA_w Num_LR_w Num_SR_w INF_w INFGrowth_w UR_
> w URGrowth_w IR_w IRGrowth_w GDP_w GDPGrowth_w i.FiscalYearCCM i.SIC_Industry 
note: 1.GFC omitted because of collinearity.
note: NONGAAPEPS omitted because of collinearity.
note: 1.COVID omitted because of collinearity.
note: NONGAAPEPS omitted because of collinearity.
note: CG omitted because of collinearity.
note: NONGAAPEPS omitted because of collinearity.
note: CG omitted because of collinearity.
note: 2009.FiscalYearCCM omitted because of collinearity.
note: 2021.FiscalYearCCM omitted because of collinearity.

      Source |       SS           df       MS      Number of obs   =    15,233
-------------+----------------------------------   F(39, 15193)    =   1178.11
       Model |  14859.2337        39  381.005992   Prob > F        =    0.0000
    Residual |   4913.4759    15,193  .323403929   R-squared       =    0.7515
-------------+----------------------------------   Adj R-squared   =    0.7509
       Total |  19772.7096    15,232  1.29810331   Root MSE        =    .56869

-----------------------------------------------------------------------------------------
EarningsPersistenceNO~w | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
------------------------+----------------------------------------------------------------
           NONGAAPEPS_w |   .8472085   .0086392    98.07   0.000     .8302746    .8641425
                     CG |   .0006837   .0038382     0.18   0.859    -.0068396    .0082069
                    GFC |  -.2457142   .1031594    -2.38   0.017    -.4479191   -.0435093
                  COVID |    .487264   .2261626     2.15   0.031     .0439581    .9305698
             NONGAAPEPS |  -.0123792   .0273156    -0.45   0.650    -.0659211    .0411626
                  1.GFC |          0  (omitted)
                        |
       GFC#c.NONGAAPEPS |
                     1  |   .0223898   .1044742     0.21   0.830    -.1823923    .2271719
                        |
             NONGAAPEPS |          0  (omitted)
                1.COVID |          0  (omitted)
                        |
     COVID#c.NONGAAPEPS |
                     1  |   .0946466   .0364757     2.59   0.009     .0231499    .1661433
                        |
             NONGAAPEPS |          0  (omitted)
                     CG |          0  (omitted)
                        |
      c.NONGAAPEPS#c.CG |   .0024844   .0022024     1.13   0.259    -.0018326    .0068013
                        |
               GFC#c.CG |
                     1  |   .0144133   .0074731     1.93   0.054    -.0002349    .0290615
                        |
  GFC#c.NONGAAPEPS#c.CG |
                     1  |  -.0153606   .0094511    -1.63   0.104    -.0338859    .0031646
                        |
             NONGAAPEPS |          0  (omitted)
                     CG |          0  (omitted)
                        |
             COVID#c.CG |
                     1  |   -.000416    .007002    -0.06   0.953    -.0141408    .0133089
                        |
COVID#c.NONGAAPEPS#c.CG |
                     1  |   -.008021   .0032677    -2.45   0.014    -.0144261    -.001616
                        |
            Num_LogTA_w |   .0149708   .0047909     3.12   0.002       .00558    .0243616
 NaturalLogarithmofFA_w |  -.0150365   .0071365    -2.11   0.035    -.0290249   -.0010482
               Num_LR_w |  -.0142566   .0046107    -3.09   0.002    -.0232941   -.0052191
               Num_SR_w |  -.0041368   .0015216    -2.72   0.007    -.0071193   -.0011543
                  INF_w |  -.0145796   .0046562    -3.13   0.002    -.0237064   -.0054528
            INFGrowth_w |    2.24604   2.484762     0.90   0.366    -2.624392    7.116473
                   UR_w |    .022627   .0131655     1.72   0.086     -.003179    .0484329
             URGrowth_w |   .0883267   .0470551     1.88   0.061    -.0039069    .1805604
                   IR_w |  -.1579794   .0214373    -7.37   0.000     -.199999   -.1159598
             IRGrowth_w |   .1500247    .046294     3.24   0.001     .0592829    .2407664
                  GDP_w |   .0000292   .0000723     0.40   0.686    -.0001125    .0001708
            GDPGrowth_w |  -.1382972    .890442    -0.16   0.877    -1.883671    1.607076
                        |
          FiscalYearCCM |
                  2007  |   .0641119   .0243402     2.63   0.008     .0164021    .1118216
                  2008  |   .0795096   .0406607     1.96   0.051    -.0001902    .1592095
                  2009  |          0  (omitted)
                  2010  |   .0378046   .0665557     0.57   0.570    -.0926526    .1682618
                  2018  |   .4377624   .1565436     2.80   0.005      .130918    .7446067
                  2019  |   .4013247   .1728043     2.32   0.020     .0626075    .7400418
                  2020  |  -.2959767   .0361518    -8.19   0.000    -.3668385   -.2251149
                  2021  |          0  (omitted)
                  2022  |   .8238776   .2370825     3.48   0.001     .3591674    1.288588
                        |
           SIC_Industry |
                     2  |   .1295933   .1024288     1.27   0.206    -.0711794    .3303661
                     3  |   .0710786   .1201032     0.59   0.554    -.1643382    .3064954
                     4  |   .1696505   .1010076     1.68   0.093    -.0283365    .3676374
                     5  |   .0885041   .1014071     0.87   0.383     -.110266    .2872742
                     6  |   .1865642   .1042873     1.79   0.074    -.0178514    .3909798
                     7  |   .1618546   .1019806     1.59   0.113    -.0380397    .3617488
                     9  |   .1393964   .1014549     1.37   0.169    -.0594674    .3382601
                    10  |   .2286225   .1189119     1.92   0.055     -.004459     .461704
                        |
                  _cons |   3.046185   .7411993     4.11   0.000     1.593345    4.499025
-----------------------------------------------------------------------------------------

. 
. eststo nongaap

. 
. 
. 
. estout

--------------------------------------
                     gaap      nongaap
                        b            b
--------------------------------------
GAAPEPS_w        .6681901             
CG                .012462     .0006837
GFC             -.0087051    -.2457142
COVID            1.715321      .487264
GAAPEPS          .0206978             
0.GFC                   0            0
1.GFC                   0            0
0.GFC#c.GA~S            0             
1.GFC#c.GA~S    -.2866625             
GAAPEPS                 0             
0.COVID                 0            0
1.COVID                 0            0
0.COVID#c.~S            0             
1.COVID#c.~S    -.0519804             
GAAPEPS                 0             
CG                      0             
c.GAAPEPS#~G    -.0072304             
0.GFC#c.CG              0            0
1.GFC#c.CG      -.0110013     .0144133
0.GFC#c.GA~G            0             
1.GFC#c.GA~G     .0168293             
GAAPEPS                 0             
CG                      0             
0.COVID#c.CG            0            0
1.COVID#c.CG    -.0255031     -.000416
0.COVID#c.~G            0             
1.COVID#c.~G     .0072397             
CG                                   0
Num_LogTA_w      .0350231     .0149708
NaturalLog~w    -.0254782    -.0150365
Num_LR_w        -.0245809    -.0142566
Num_SR_w        -.0028228    -.0041368
INF_w           -.0152702    -.0145796
INFGrowth_w      9.317135      2.24604
UR_w             .0304478      .022627
URGrowth_w      -.0034263     .0883267
IR_w            -.1400874    -.1579794
IRGrowth_w       .1033606     .1500247
GDP_w           -.0001371     .0000292
GDPGrowth_w     -.2230232    -.1382972
2006.F~arCCM            0            0
2007.F~arCCM       .10874     .0641119
2008.F~arCCM     .0594729     .0795096
2009.F~arCCM            0            0
2010.F~arCCM     .0831276     .0378046
2018.F~arCCM     1.206829     .4377624
2019.F~arCCM     1.169862     .4013247
2020.F~arCCM    -.4886415    -.2959767
2021.F~arCCM            0            0
2022.F~arCCM     1.851714     .8238776
1.SIC_Indu~y            0            0
2.SIC_Indu~y     .0953184     .1295933
3.SIC_Indu~y      .002676     .0710786
4.SIC_Indu~y     .2374448     .1696505
5.SIC_Indu~y     .0942036     .0885041
6.SIC_Indu~y     .2452833     .1865642
7.SIC_Indu~y     .2960358     .1618546
9.SIC_Indu~y     .1679514     .1393964
10.SIC_Ind~y     .4052408     .2286225
NONGAAPEPS_w                  .8472085
NONGAAPEPS                   -.0123792
0.GFC#c.NO~S                         0
1.GFC#c.NO~S                  .0223898
NONGAAPEPS                           0
0.COVID#c.~S                         0
1.COVID#c.~S                  .0946466
NONGAAPEPS                           0
NONGAAPEPS                           0
c.NONGAAPE~G                  .0024844
0.GFC#c.NO~G                         0
1.GFC#c.NO~G                 -.0153606
CG                                   0
0.COVID#c.~G                         0
1.COVID#c.~G                  -.008021
_cons            5.628291     3.046185
--------------------------------------

. 
. 
. 
. suest gaap nongaap, vce(robust)

Simultaneous results for gaap, nongaap                  Number of obs = 15,233

-----------------------------------------------------------------------------------------
                        |               Robust
                        | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------------+----------------------------------------------------------------
gaap_mean               |
              GAAPEPS_w |   .6681901   .0351235    19.02   0.000     .5993492     .737031
                     CG |    .012462   .0087594     1.42   0.155    -.0047062    .0296302
                    GFC |  -.0087051   .1819251    -0.05   0.962    -.3652718    .3478616
                  COVID |   1.715321   .5058938     3.39   0.001     .7237874    2.706855
                GAAPEPS |   .0206978   .0708347     0.29   0.770    -.1181356    .1595311
                  1.GFC |          0  (omitted)
                        |
          GFC#c.GAAPEPS |
                     1  |  -.2866625   .1513822    -1.89   0.058    -.5833662    .0100412
                        |
                GAAPEPS |          0  (omitted)
                1.COVID |          0  (omitted)
                        |
        COVID#c.GAAPEPS |
                     1  |  -.0519804   .0995265    -0.52   0.601    -.2470488    .1430879
                        |
                GAAPEPS |          0  (omitted)
                     CG |          0  (omitted)
                        |
         c.GAAPEPS#c.CG |  -.0072304   .0074576    -0.97   0.332    -.0218471    .0073862
                        |
               GFC#c.CG |
                     1  |  -.0110013   .0113085    -0.97   0.331    -.0331656     .011163
                        |
     GFC#c.GAAPEPS#c.CG |
                     1  |   .0168293   .0132493     1.27   0.204    -.0091388    .0427974
                        |
                GAAPEPS |          0  (omitted)
                     CG |          0  (omitted)
                        |
             COVID#c.CG |
                     1  |  -.0255031   .0160048    -1.59   0.111     -.056872    .0058658
                        |
   COVID#c.GAAPEPS#c.CG |
                     1  |   .0072397    .009761     0.74   0.458    -.0118914    .0263709
                        |
            Num_LogTA_w |   .0350231   .0092024     3.81   0.000     .0169867    .0530595
 NaturalLogarithmofFA_w |  -.0254782   .0139197    -1.83   0.067    -.0527604     .001804
               Num_LR_w |  -.0245809   .0076462    -3.21   0.001    -.0395672   -.0095947
               Num_SR_w |  -.0028228    .003259    -0.87   0.386    -.0092102    .0035647
                  INF_w |  -.0152702   .0086821    -1.76   0.079    -.0322867    .0017463
            INFGrowth_w |   9.317135   4.090583     2.28   0.023      1.29974    17.33453
                   UR_w |   .0304478   .0210444     1.45   0.148    -.0107985    .0716941
             URGrowth_w |  -.0034263   .0816399    -0.04   0.967    -.1634376     .156585
                   IR_w |  -.1400874   .0346359    -4.04   0.000    -.2079726   -.0722022
             IRGrowth_w |   .1033606   .0793562     1.30   0.193    -.0521746    .2588959
                  GDP_w |  -.0001371   .0001126    -1.22   0.223    -.0003578    .0000836
            GDPGrowth_w |  -.2230232   1.625973    -0.14   0.891    -3.409872    2.963826
                        |
          FiscalYearCCM |
                  2007  |     .10874    .035793     3.04   0.002     .0385871    .1788929
                  2008  |   .0594729    .074722     0.80   0.426    -.0869796    .2059254
                  2009  |          0  (omitted)
                  2010  |   .0831276   .1343013     0.62   0.536    -.1800981    .3463534
                  2018  |   1.206829   .3430556     3.52   0.000     .5344525    1.879206
                  2019  |   1.169862   .3756789     3.11   0.002     .4335447    1.906179
                  2020  |  -.4886415   .0764388    -6.39   0.000    -.6384589   -.3388242
                  2021  |          0  (omitted)
                  2022  |   1.851714    .530377     3.49   0.000     .8121938    2.891233
                        |
           SIC_Industry |
                     2  |   .0953184   .1706337     0.56   0.576    -.2391175    .4297542
                     3  |    .002676    .184242     0.01   0.988    -.3584317    .3637836
                     4  |   .2374448    .164188     1.45   0.148    -.0843577    .5592473
                     5  |   .0942036   .1652241     0.57   0.569    -.2296298    .4180369
                     6  |   .2452833   .1728972     1.42   0.156    -.0935889    .5841555
                     7  |   .2960358   .1664092     1.78   0.075    -.0301202    .6221918
                     9  |   .1679514    .164978     1.02   0.309    -.1553995    .4913024
                    10  |   .4052408   .2009133     2.02   0.044     .0114579    .7990237
                        |
                  _cons |   5.628291    1.30525     4.31   0.000     3.070048    8.186534
------------------------+----------------------------------------------------------------
gaap_lnvar              |
                  _cons |   .1042814   .0330921     3.15   0.002     .0394221    .1691408
------------------------+----------------------------------------------------------------
nongaap_mean            |
           NONGAAPEPS_w |   .8472085   .0158079    53.59   0.000     .8162256    .8781915
                     CG |   .0006837   .0034654     0.20   0.844    -.0061083    .0074757
                    GFC |  -.2457142   .1005351    -2.44   0.015    -.4427594    -.048669
                  COVID |    .487264   .2802374     1.74   0.082    -.0619911    1.036519
             NONGAAPEPS |  -.0123792   .0408702    -0.30   0.762    -.0924833    .0677248
                  1.GFC |          0  (omitted)
                        |
       GFC#c.NONGAAPEPS |
                     1  |   .0223898    .139456     0.16   0.872    -.2509388    .2957184
                        |
             NONGAAPEPS |          0  (omitted)
                1.COVID |          0  (omitted)
                        |
     COVID#c.NONGAAPEPS |
                     1  |   .0946466   .0605949     1.56   0.118    -.0241172    .2134104
                        |
             NONGAAPEPS |          0  (omitted)
                     CG |          0  (omitted)
                        |
      c.NONGAAPEPS#c.CG |   .0024844    .003295     0.75   0.451    -.0039736    .0089424
                        |
               GFC#c.CG |
                     1  |   .0144133   .0072461     1.99   0.047     .0002113    .0286153
                        |
  GFC#c.NONGAAPEPS#c.CG |
                     1  |  -.0153606   .0133801    -1.15   0.251    -.0415851    .0108639
                        |
             NONGAAPEPS |          0  (omitted)
                     CG |          0  (omitted)
                        |
             COVID#c.CG |
                     1  |   -.000416   .0086625    -0.05   0.962    -.0173941    .0165621
                        |
COVID#c.NONGAAPEPS#c.CG |
                     1  |   -.008021   .0058528    -1.37   0.171    -.0194924    .0034503
                        |
            Num_LogTA_w |   .0149708    .004744     3.16   0.002     .0056727    .0242689
 NaturalLogarithmofFA_w |  -.0150365   .0071397    -2.11   0.035    -.0290302   -.0010429
               Num_LR_w |  -.0142566   .0039344    -3.62   0.000    -.0219678   -.0065454
               Num_SR_w |  -.0041368   .0019611    -2.11   0.035    -.0079805   -.0002931
                  INF_w |  -.0145796   .0047466    -3.07   0.002    -.0238828   -.0052764
            INFGrowth_w |    2.24604   2.358728     0.95   0.341    -2.376982    6.869063
                   UR_w |    .022627   .0119804     1.89   0.059    -.0008542    .0461082
             URGrowth_w |   .0883267   .0455381     1.94   0.052    -.0009263    .1775797
                   IR_w |  -.1579794   .0199646    -7.91   0.000    -.1971093   -.1188494
             IRGrowth_w |   .1500247   .0445229     3.37   0.001     .0627614     .237288
                  GDP_w |   .0000292   .0000616     0.47   0.635    -.0000915    .0001498
            GDPGrowth_w |  -.1382972   .8950393    -0.15   0.877    -1.892542    1.615948
                        |
          FiscalYearCCM |
                  2007  |   .0641119   .0195303     3.28   0.001     .0258331    .1023906
                  2008  |   .0795096   .0396725     2.00   0.045     .0017531    .1572662
                  2009  |          0  (omitted)
                  2010  |   .0378046   .0746341     0.51   0.612    -.1084755    .1840847
                  2018  |   .4377624   .1914529     2.29   0.022     .0625216    .8130032
                  2019  |   .4013247   .2119479     1.89   0.058    -.0140857     .816735
                  2020  |  -.2959767    .044296    -6.68   0.000    -.3827954   -.2091581
                  2021  |          0  (omitted)
                  2022  |   .8238776   .2983486     2.76   0.006     .2391251     1.40863
                        |
           SIC_Industry |
                     2  |   .1295933   .1699649     0.76   0.446    -.2035318    .4627184
                     3  |   .0710786   .1757142     0.40   0.686    -.2733149     .415472
                     4  |   .1696505   .1691783     1.00   0.316     -.161933    .5012339
                     5  |   .0885041   .1696316     0.52   0.602    -.2439678     .420976
                     6  |   .1865642   .1705636     1.09   0.274    -.1477343    .5208626
                     7  |   .1618546   .1701995     0.95   0.342    -.1717304    .4954395
                     9  |   .1393964   .1695515     0.82   0.411    -.1929184    .4717111
                    10  |   .2286225   .1716021     1.33   0.183    -.1077115    .5649565
                        |
                  _cons |   3.046185   .7501229     4.06   0.000     1.575971    4.516399
------------------------+----------------------------------------------------------------
nongaap_lnvar           |
                  _cons |  -1.128853   .0418945   -26.95   0.000    -1.210965   -1.046742
-----------------------------------------------------------------------------------------

. 
. test [gaap_mean]GAAPEPS_w=[nongaap_mean]NONGAAPEPS_w

 ( 1)  [gaap_mean]GAAPEPS_w - [nongaap_mean]NONGAAPEPS_w = 0

           chi2(  1) =   27.00
         Prob > chi2 =    0.0000

. 
. test [gaap_mean]GFC=[nongaap_mean]GFC

 ( 1)  [gaap_mean]GFC - [nongaap_mean]GFC = 0

           chi2(  1) =    1.92
         Prob > chi2 =    0.1661

. 
. test [gaap_mean]COVID=[nongaap_mean]COVID

 ( 1)  [gaap_mean]COVID - [nongaap_mean]COVID = 0

           chi2(  1) =    8.47
         Prob > chi2 =    0.0036

. 
. test [gaap_mean]c.GAAPEPS##i.GFC=[nongaap_mean]c.NONGAAPEPS##i.GFC
invalid matrix stripe;
c.GAAPEPS##i.GFC
r(198);

. 
. test [gaap_mean]c.GAAPEPS##i.COVID=[nongaap_mean]c.NONGAAPEPS##i.COVID
invalid matrix stripe;
c.GAAPEPS##i.COVID
r(198);

. 
. test [gaap_mean]c.GAAPEPS##i.GFC##c.CG=[nongaap_mean]c.NONGAAPEPS##i.GFC##c.CG
invalid matrix stripe;
c.GAAPEPS##i.GFC##c.CG
r(198);

. 
. test [gaap_mean]c.GAAPEPS##i.COVID##c.CG=[nongaap_mean]c.NONGAAPEPS##i.COVID##c.CG
invalid matrix stripe;
c.GAAPEPS##i.COVID##c.CG
r(198);

[Clyde Schechter]

Will i.year and i.industry capture the fixed effects in the correct way?

Yes.

Concerning the "invalid matrix stripe" errors, they arise because you cannot use i.whatever in this context. Each of the things that appear in a [gap_mean]something context must be the actual name of a coefficient calculated by -suest-, not the name of a term in the underlying regression command(s). Getting these names right can be tricky, as different commands make them up differently, and even the same command can do it differently in different Stata versions. So, rerun your -suest- command, but this time at the -coeflegend- command. The output will give you the same coefficients you got before, but instead of the associated inferential statistics, you will get the exact names of the coefficients. Use those in your -test- commands to avoid these error messages.

[Jan Arendszoon]

Thanks Clyde, I truly appreciate the help. Have a good day!