Table 5: Proportion of 5000 simulations with B = 1000 bootstrap draws for each simulation where future subjects with covariates x 0 are denied treatment 1 for a significant difference within subgroups (column 3), an ITE estimator greater than c (column 4) or a probability of at least 0.5 for the ITE estimator to be greater than c (column 5) with c = log (1.2). The last row shows the results for a prediction model that is derived from 1000 subjects per treatment with x 1 = x 2 = 0 instead of 100 while all the other settings remain unchanged.

x 01 x 02 Pr[ sig diff ] Pr[ D ^ ( x 0 )   c ] Pr[ P ^ * ( x 0 , c )   0.5 ]
1 1 0.3184 0.8072 0.8068
1 0 0.1498 0.630 0.6304
0 1 0.0912 0.4924 0.4894
0 0 0.0692 0.3714 0.3652
0 0 0.0958 0.1864 0.1872