Parallel trends regression test for differnce in difference

[Anuradha Saikia]

Hello everyone. I am running a difference in difference on a set of firms before and after a reform act in 2014. The treatment is set of firms who are spending on CSR post 2013 reform act . The treatment group is Sp_D .
however i get the following result. I am not sure how to deal with this. Please help

#

 reghdfe ROA i.Sp_D##ib(2013).Year, absorb( Companynum Year) vce(cl Companynum )
note: 1bn.Sp_D is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2005bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2006bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2007bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2008bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2009bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2010bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2011bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2012bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2014bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2015bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2016bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2017bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2018bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2019bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
note: 2020bn.Year is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0e-09)
(MWFE estimator converged in 5 iterations)
note: 1.Sp_D omitted because of collinearity
note: 2005.Year omitted because of collinearity
note: 2006.Year omitted because of collinearity
note: 2007.Year omitted because of collinearity
note: 2008.Year omitted because of collinearity
note: 2009.Year omitted because of collinearity
note: 2010.Year omitted because of collinearity
note: 2011.Year omitted because of collinearity
note: 2012.Year omitted because of collinearity
note: 2014.Year omitted because of collinearity
note: 2015.Year omitted because of collinearity
note: 2016.Year omitted because of collinearity
note: 2017.Year omitted because of collinearity
note: 2018.Year omitted because of collinearity
note: 2019.Year omitted because of collinearity
note: 2020.Year omitted because of collinearity

HDFE Linear regression                            Number of obs   =      2,687
Absorbing 2 HDFE groups                           F(  15,    174) =       0.73
Statistics robust to heteroskedasticity           Prob > F        =     0.7480
                                                  R-squared       =     0.0776
                                                  Adj R-squared   =     0.0014
                                                  Within R-sq.    =     0.0060
Number of clusters (Companynum) =        175      Root MSE        =  2752.9112

                           (Std. Err. adjusted for 175 clusters in Companynum)
------------------------------------------------------------------------------
             |               Robust
         ROA |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      1.Sp_D |          0  (omitted)
             |
        Year |
       2005  |          0  (omitted)
       2006  |          0  (omitted)
       2007  |          0  (omitted)
       2008  |          0  (omitted)
       2009  |          0  (omitted)
       2010  |          0  (omitted)
       2011  |          0  (omitted)
       2012  |          0  (omitted)
       2014  |          0  (omitted)
       2015  |          0  (omitted)
       2016  |          0  (omitted)
       2017  |          0  (omitted)
       2018  |          0  (omitted)
       2019  |          0  (omitted)
       2020  |          0  (omitted)
             |
   Sp_D#Year |
     1 2005  |  -142.5125   133.0878    -1.07   0.286    -405.1867    120.1618
     1 2006  |   2.473112   37.69682     0.07   0.948    -71.92879    76.87501
     1 2007  |  -1756.098   1741.915    -1.01   0.315      -5194.1    1681.904
     1 2008  |  -11.88468    11.2148    -1.06   0.291    -34.01923    10.24988
     1 2009  |   31.82636   33.83661     0.94   0.348    -34.95668    98.60939
     1 2010  |    4.23284   9.794813     0.43   0.666     -15.0991    23.56478
     1 2011  |    -2.2177   5.168096    -0.43   0.668    -12.41793    7.982528
     1 2012  |  -.6721107   5.578268    -0.12   0.904    -11.68189    10.33767
     1 2014  |  -4.665616   3.583425    -1.30   0.195    -11.73819    2.406959
     1 2015  |    -4.0877   3.793364    -1.08   0.283    -11.57463     3.39923
     1 2016  |  -5.637149   3.886853    -1.45   0.149     -13.3086    2.034299
     1 2017  |   1.270018   5.777267     0.22   0.826    -10.13252    12.67256
     1 2018  |   62.26375   67.62298     0.92   0.358    -71.20316    195.7306
     1 2019  |  -.9259057   7.921466    -0.12   0.907    -16.56044    14.70862
     1 2020  |  -9.488381   4.617248    -2.05   0.041    -18.60141   -.3753577
             |
       _cons |   107.0989   51.62026     2.07   0.039     5.216483    208.9814
------------------------------------------------------------------------------

Absorbed degrees of freedom:
-----------------------------------------------------+
 Absorbed FE | Categories  - Redundant  = Num. Coefs |
-------------+---------------------------------------|
  Companynum |       175         175           0    *|
        Year |        16           0          16     |
-----------------------------------------------------+
* = FE nested within cluster; treated as redundant for DoF computation

[Jared Greathouse]

It looks like (i can't tell, you don't provide data as we specifically ask in the FAQ) your treatment variable is always 1.


This has nothing to do with parallel trends, we would expect the treatment to be perfectly collinear with year effects in this case, so this is likely the reason your year effects drop out. You need your treated units to be =to 0 before they were treated, otherwise this will happen

[Anuradha Saikia]

Thanks Jared Greathouse . I understood where my mistake is coming from.