Xtdcce_m_reg (): 3200 conformability error

[Firmin Clairant]

Hi Dear Member,


In fact, I'm trying to use your program (xtdcce) to estimate a model and I get this error:


"Xtdcce_m_reg (): 3200 conformability error". Any idea what does that mean? Could it be because the panel is unbalanced and we are having too many coefficients to be estimated?


Best regards,

[Elkhan Richard Sadik-Zada]

no, this is definitely not the case. I have a model with only two variables and get the same message...

[JanDitzen]

Can you please give me some more info about the dataset you use, the command line you use for xtdcce2 and the version of xtdcce2 and Stata. Please note, that version 1.34 is available on my website (https://janditzen.github.io/xtdcce2/) and can be installed in Stata:

net install xtdcce2 , from("https://janditzen.github.io/xtdcce2/")

or

net from https://janditzen.github.io/xtdcce2/

[Atif Awad]

Dear Jan , Greeting

I got the same error
I have balnaced panel data for 47 countries (1990-2017).

xtdcce_m_meangroup(): 3200 conformability error
<istmt>: - function returned error
r(3200);

When I implemented your recommendation above, i got the below

net install xtdcce2 , from("https://janditzen.github.io/xtdcce2/")
checking xtdcce2 consistency and verifying not already installed...

the following files already exist and are different:
c:\ado\plus\x\xtdcce2.ado
c:\ado\plus\x\xtdcce2.sthlp
c:\ado\plus\x\xtset2.sthlp
c:\ado\plus\x\xtcd2.ado
c:\ado\plus\x\xtcd2.sthlp
c:\ado\plus\x\xtcse2.sthlp

no files installed or copied
(no action taken)
r(602);

I would appreciate
your help

[JanDitzen]

Dear Atif,
can you please install the beta version of the next xtdcce2 version. The version is called xtdcce221 and can be installed by typing:

net install xtdcce221 , from("https://janditzen.github.io/xtdcce2/")

This should fix the error.

Hope that helps.

Best

[Atif Awad]

Dear Jan
Many thanks, problem solved.

[Atif Awad]

Dear Jan, Happy new year
do you know I tried using your last suggestion, but I received

. net install xtdcce221 , from("https://janditzen.github.io/xtdcce2/")
Java installation not found
https://janditzen.github.io/xtdcce2/ either
1) is not a valid URL, or
2) could not be contacted, or
3) is not a Stata download site (has no stata.toc file).
r(5004);

. net from https://janditzen.github.io/xtdcce2/
Java installation not found
https://janditzen.github.io/xtdcce2/ either
1) is not a valid URL, or
2) could not be contacted, or
3) is not a Stata download site (has no stata.toc file).
r(5004);

I don't know how to solve this matter?

[JanDitzen]

Hi Atif,
the Java problem seems to be computer specific. Can you please check if you have the latest version of Java installed?

In the meantime, you can use the version from SSC which is currently version 3.0, so it superseeds version 2.21.

Best,

Jan

[Marco Astori]

I've the same error

 
Units (Code) to be removed due to insufficient numbers of observations:  93 108
    xtdcce_m_meangroup():  3200  conformability error
                 <istmt>:     -  function returned error
r(3200);

JanDitzen how can I solve it?

[JanDitzen]

Hi Marco,
which version of xtdcce2 do you use? You can install the latest version of xtdcce2 (v. 3) from either SSC or from GitHub:

 
 net install xtdcce2 , from("https://janditzen.github.io/xtdcce2/")

Can you also please provide the 1) command line you use, 2) number of time periods and cross-sections and if the dataset is balanced or not.

Thanks,

Jan

[Marco Astori]

I've installed the version from GitHub with

  
 net install xtdcce2 , from("https://janditzen.github.io/xtdcce2/")

The answer has been

 checking xtdcce2 consistency and verifying not already installed...  the following files will be replaced:     C:\Users\nico4\ado\plus\x\xtset2.sthlp     C:\Users\nico4\ado\plus\x\xtdcce2_auxiliary.ado  installing into C:\Users\nico4\ado\plus\... installation complete.

My panel is strongly balanced, I've less than 20 periods of time and 117 cross sections. The command is

xtdcce2 LSECI L.LSECI LRR1 LP LSFI1 (L.LSECI= L3.LSECI), crosssectional(L.LSECI LRR1 LP LSFI1) jackknife

[JanDitzen]

Thanks,
What happens if you do not use the jackknife option, does it still crash?

From the initial output it looked like some observations have missing values. Is this true?

[Marco Astori]

JanDitzen I've missing values and without the jackknife the error is

 
. xtdcce2 LSECI L.LSECI LRR1 LP LSFI1 (L.LSECI= L3.LSECI), crosssectional(L.LSECI LRR1 LP LSFI1)
Units (Code) to be removed due to insufficient numbers of observations:  93 108
    xtdcce_m_meangroup():  3200  conformability error
                 <istmt>:     -  function returned error
r(3200);

[JanDitzen]

This means that your panel is effectively unbalanced. xtdcce2 should detect this.

I noted that in the syntax L.LSECI appears twice. Can you please remove one of those. Does this help?

[Marco Astori]

Yes it helps, codes are produced.

 
. xtdcce2 LSECI L.LSECI LRR1 LP LSFI1 , crosssectional(L.LSECI LRR1 LP LSFI1) jackknife
Units (Code) to be removed due to insufficient numbers of observations:  93
(Dynamic) Common Correlated Effects Estimator - Mean Group

Panel Variable (i): Code                         Number of obs     =       1788
Time Variable (t): year                          Number of groups  =        112

Degrees of freedom per group:                    Obs per group:    
 without cross-sectional avg. min   = 7                        min =         12
                              max   = 11                       avg =         16
 with cross-sectional avg.    min   = 3                        max =         16
                              max   = 7
Number of                                        F(1008, 780)      =       0.77
 cross-sectional lags               0 to 0       Prob > F          =       1.00
 variables in mean group regression = 448        R-squared         =   -5100.35
 variables partialled out           = 560        R-squared (MG)    =   -7059.46
                                                 Root MSE          =      20.32
                                                 CD Statistic      =      -0.08
                                                    p-value        =     0.9374
-------------------------------------------------------------------------------
          LSECI|     Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
---------------+---------------------------------------------------------------
 Mean Group:   |
        L.LSECI| -.4865558   .4056879   -1.20    0.230      -1.28169    .308578
           LRR1| -.7788234   2.080121   -0.37    0.708     -4.855786   3.298139
             LP|  240.0987   176.8949    1.36    0.175     -106.6089   586.8063
          LSFI1| -.0210393   .3540347   -0.06    0.953     -.7149346   .6728561
-------------------------------------------------------------------------------
Mean Group Variables: L.LSECI LRR1 LP LSFI1
Cross Sectional Averaged Variables: L.LSECI LRR1 LP LSFI1
Heterogenous constant partialled out. 
Jackknife bias correction used. 
Warning:
Rank condition on matrix of cross product of cross sectional averages not satisfied!
Only mean group estimates are consistent, unit specific estimates are inconsistent.
Warning:
Collinearities detected. One or more variables are dropped and set to zero.
Use estat ebistructure to display more details.

Is L.LSECI in this case endogenous? Because of correlation with fixed effects.