st: Comformability error with -margins-

[Suryadipta Roy <sroy2138@gmail.com>]

Dear Statalisters,

I am having some problems with -margins- and would very much
appreciate if someone can help me out. The model is as follows:
areg lnimport1 i.importer_gatt##c.icrgm control_variables,
absorb(country) robust, where importer_gatt is categorical variable
(0/1) and icrgm is a continuous variable (0 - 6).

The results with -margins- is below and this will highlight the problem:

margins importer_gatt, at(icrgm=0) noestimcheck  /* THIS WORKS */

Predictive margins                                Number of obs   =     114749
Model VCE    : Robust

Expression   : Linear prediction, predict()
at           : icrgm           =           0

------------------------------------------------------------------------------
             |            Delta-method
             |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
importer_g~t |
          0  |   12.18997   .2452575    49.70   0.000     11.70928    12.67067
          1  |   14.28279   .0813169   175.64   0.000     14.12341    14.44216
------------------------------------------------------------------------------

margins r.importer_gatt, at(icrgm=0) noestimcheck /* THIS ALSO WORKS */

Contrasts of predictive margins
Model VCE    : Robust

Expression   : Linear prediction, predict()
at           : icrgm           =           0

-------------------------------------------------
              |         df        chi2     P>chi2
--------------+----------------------------------
importer_gatt |          1       65.93     0.0000
-------------------------------------------------

---------------------------------------------------------------
              |            Delta-method
              |   Contrast   Std. Err.     [95% Conf. Interval]
--------------+------------------------------------------------
importer_gatt |
    (1 vs 0)  |   2.092815   .2577455      1.587643    2.597986
---------------------------------------------------------------

/* BUT THIS DOES NOT WORK */
margins r.importer_gatt, at(icrgm=(0 6)) noestimcheck
conformability error
r(503);

In other words, I am getting a message for Conformability error when I
try to compare the adjusted means for different values of icrgm. The
above does not work without the r. contrast operator as well. The
weird part is that any combination of the above code has worked before
for me (even this morning)!

The .do file also consists of the following codes before (the problem
appears) that run alright.

anova lnimport1 honest_imp##importer_gatt; /* honest_imp is a (0/1)
categorical variable*/
margins importer_gatt#honest_imp;
marginsplot, noci;
contrast importer_gatt@honest_imp, pveffects nowald;
contrast importer_gatt#honest_imp, pveffects nowald;

Any help is greatly appreciated.

Sincerely,
Suryadipta Roy.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/