Skip to main content

Table 2 Means and standard deviations of the estimates of C for a dichotomized variable

From: Polychotomization of continuous variables in regression models based on the overall C index

  true C
n 0.650 0.750 0.850
10 0.662 ± 0.079 0.760 ± 0.076 0.856 ± 0.062
25 0.655 ± 0.051 0.754 ± 0.048 0.852 ± 0.040
50 0.652 ± 0.036 0.751 ± 0.034 0.851 ± 0.028
100 0.651 ± 0.026 0.751 ± 0.024 0.851 ± 0.020
200 0.650 ± 0.018 0.750 ± 0.017 0.850 ± 0.014
500 0.650 ± 0.011 0.750 ± 0.011 0.850 ± 0.009
  1. Two samples of the size n from N(0, 12) and N(μ d , 1.52) are generated by Monte Carlo simulation and the pair consistency probability C for the optimal dichotomized point is calculated by the parametric method. This step is iterated 10,000 times, producing means and standard deviations of C. The values of μ d are 0.840, 1.613 and 2.543 for the true values of C 0.650, 0.750 and 0.850, respectively.