Hello,

I am running a logistic regression with the log odds and the marginal effects as coefficients. However, stata returns different levels of significance. Could anyone clarify how this comes?

Code:
. logit change firmsize profitability leverage age capitalintensity CAPEX KZindex elektricityge
> nerator Carbonleakage SME publicfirm, robust

Iteration 0:   log pseudolikelihood = -115.72863  
Iteration 1:   log pseudolikelihood = -104.96108  
Iteration 2:   log pseudolikelihood = -104.95884  
Iteration 3:   log pseudolikelihood = -104.95884  

Logistic regression                             Number of obs     =        167
                                                Wald chi2(11)     =      20.39
                                                Prob > chi2       =     0.0403
Log pseudolikelihood = -104.95884               Pseudo R2         =     0.0931

--------------------------------------------------------------------------------------
                     |               Robust
              change |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
            firmsize |   .1474855   .1479958     1.00   0.319    -.1425809    .4375518
       profitability |  -4.407245   5.478104    -0.80   0.421    -15.14413    6.329641
            leverage |  -1.341288     .76225    -1.76   0.078     -2.83527    .1526947
                 age |   .0134197   .0080987     1.66   0.098    -.0024536    .0292929
    capitalintensity |   -1.84617   .8252756    -2.24   0.025     -3.46368   -.2286593
               CAPEX |   1.984933   2.759231     0.72   0.472    -3.423061    7.392927
             KZindex |    .114081   .1276615     0.89   0.372    -.1361311     .364293
elektricitygenerator |  -1.031212   1.289364    -0.80   0.424    -3.558318    1.495894
       Carbonleakage |  -.5588379   .3587584    -1.56   0.119    -1.261991    .1443156
                 SME |   1.341953   .5178452     2.59   0.010     .3269952    2.356911
          publicfirm |   .2272293    1.13111     0.20   0.841    -1.989706    2.444165
               _cons |  -2.104979   2.895328    -0.73   0.467    -7.779718    3.569761

. margins, dydx(firmsize profitability leverage age capitalintensity CAPEX KZindex elektricityg
> enerator Carbonleakage SME publicfirm)

Average marginal effects                        Number of obs     =        167
Model VCE    : Robust

Expression   : Pr(change), predict()
dy/dx w.r.t. : firmsize profitability leverage age capitalintensity CAPEX KZindex
               elektricitygenerator Carbonleakage SME publicfirm

--------------------------------------------------------------------------------------
                     |            Delta-method
                     |      dy/dx   Std. Err.      z    P>|z|     [95% Conf. Interval]
---------------------+----------------------------------------------------------------
            firmsize |   .0323665   .0321335     1.01   0.314     -.030614    .0953469
       profitability |  -.9671929   1.190567    -0.81   0.417    -3.300661    1.366275
            leverage |  -.2943526   .1623227    -1.81   0.070    -.6124993    .0237941
                 age |    .002945   .0017276     1.70   0.088     -.000441     .006331
    capitalintensity |  -.4051515    .171228    -2.37   0.018    -.7407523   -.0695508
               CAPEX |   .4356038   .6013645     0.72   0.469     -.743049    1.614257
             KZindex |   .0250357   .0279073     0.90   0.370    -.0296616    .0797329
elektricitygenerator |  -.2263049   .2807979    -0.81   0.420    -.7766587     .324049
       Carbonleakage |  -.1226399   .0774626    -1.58   0.113    -.2744639    .0291841
                 SME |   .2944986    .106381     2.77   0.006     .0859957    .5030015
          publicfirm |   .0498666   .2479409     0.20   0.841    -.4360885    .5358218
kind regards,
Timea