gistfile1.txt

. do "/var/folders/47/f2b3zfds1y5gq4vp7m6___cr0000gn/T//SD44048.000000"

. clear all

. set seed 123

. 
. * recruit subjects (if only this easy...)
. set obs 2000
number of observations (_N) was 0, now 2,000

. 
. * assign to groups
. gen mbct =  mod(_n, 2)

. 
. * depression baseline score
. gen d0 = rnormal(0,1)

. 
. * generate probability of relapse note - no interaction term here
. gen p = 1 - min(.5*d0 + .3*mbct, 1) 

. 
. summ

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
        mbct |      2,000          .5     .500125          0          1
          d0 |      2,000    .0203836    .9839001  -4.604603    3.40045
           p |      2,000     .851266    .4890847          0   3.302301

. codebook

------------------------------------------------------------------------------------------------
mbct                                                                                 (unlabeled)
------------------------------------------------------------------------------------------------

                  type:  numeric (float)

                 range:  [0,1]                        units:  1
         unique values:  2                        missing .:  0/2,000

            tabulation:  Freq.  Value
                         1,000  0
                         1,000  1

------------------------------------------------------------------------------------------------
d0                                                                                   (unlabeled)
------------------------------------------------------------------------------------------------

                  type:  numeric (float)

                 range:  [-4.6046028,3.4004498]       units:  1.000e-11
         unique values:  2,000                    missing .:  0/2,000

                  mean:   .020384
              std. dev:     .9839

           percentiles:        10%       25%       50%       75%       90%
                          -1.21403  -.612031   .012687   .663429   1.23962

------------------------------------------------------------------------------------------------
p                                                                                    (unlabeled)
------------------------------------------------------------------------------------------------

                  type:  numeric (float)

                 range:  [0,3.3023014]                units:  1.000e-10
         unique values:  1,901                    missing .:  0/2,000

                  mean:   .851266
              std. dev:   .489085

           percentiles:        10%       25%       50%       75%       90%
                           .179416   .496526   .840672   1.17163   1.50491

. 
. gen y = rbinomial(1,p)
(843 missing values generated)

. 
. * ns interaction term in the logistic regression
. logistic y mbct##c.d0

Logistic regression                             Number of obs     =      1,157
                                                LR chi2(3)        =     379.01
                                                Prob > chi2       =     0.0000
Log likelihood = -596.29243                     Pseudo R2         =     0.2412

------------------------------------------------------------------------------
           y | Odds Ratio   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      1.mbct |   .2666635   .0682103    -5.17   0.000     .1615226    .4402443
          d0 |   .0737559   .0198726    -9.68   0.000     .0434962    .1250667
             |
   mbct#c.d0 |
          1  |    .919064   .3135055    -0.25   0.805     .4709655    1.793504
             |
       _cons |   10.79623   2.467407    10.41   0.000     6.898198    16.89696
------------------------------------------------------------------------------

. 
. * but get ES estimates in probability domainfo 1 SD above and below mean baseline 
. margins, dydx(mbct) at(d0=(-1(1)1)) 

Conditional marginal effects                    Number of obs     =      1,157
Model VCE    : OIM

Expression   : Pr(y), predict()
dy/dx w.r.t. : 1.mbct

1._at        : d0              =          -1

2._at        : d0              =           0

3._at        : d0              =           1

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
1.mbct       |
         _at |
          1  |  -.0162185   .0072951    -2.22   0.026    -.0305166   -.0019204
          2  |  -.1730282   .0282415    -6.13   0.000    -.2283805   -.1176759
          3  |  -.2800077   .0394602    -7.10   0.000    -.3573483   -.2026671
------------------------------------------------------------------------------
Note: dy/dx for factor levels is the discrete change from the base level.

. 
.