Skip to main content

Table 4 Simulation results in a domain with ICC = 15%, Pearson correlation X1 and random effect 0 .0

From: Prediction models for clustered data: comparison of a random intercept and standard regression model

  Apparent performance Test performance
  Standard model Marginal risk calculation Conditional risk calculation Standard model Marginal risk calculation
Harrell’s C-index † 0.77 (0.735; 0.802) 0.77 (0.734; 0.801) 0.85 (0.818; 0.873) 0.77 (0.762; 0.769) 0.77 (0.763; 0.769)
C-index within clusters ‡ 0.80 (0.085) 0.80 (0.086) 0.80 (0.086) 0.80 (0.037) 0.80 (0.037)
Calibration intercept* 0.00 0.19 0.00 -0.00 0.20
Calibration intercept* within clusters ‡ -0.18 (0.941) -0.00 (1.005) 0.00 (0.000) -0.19 (0.967) 0.01 (1.024)
Calibration slope 1.00 0.85 1.07 0.97 0.83
Calibration slope within clusters ‡ 1.18 (0.096) 1.00 (0.082) 1.07 (0.000) 1.15 (0.011) 0.99 (0.008)
  1. * With calibration slope equal to 1 (i.e. calibration in the large).
  2. † overall performance (2.5 and 97.5 percentiles).
  3. ‡ median of overall performance from 100 simulations (median of within cluster performances).