Skip to main content

Table 3 Simulation results in a domain with ICC = 5%, 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 †

0.79 (0.766; 0.816)

0.79 (0.766; 0.816)

0.82 (0.788; 0.839)

0.78 (0.780; 0.785)

0.78 (0.780; 0.785)

C-index within clusters ‡

0.80 (0.077)

0.80 (0.077)

0.80 (0.077)

0.79 (0.031)

0.79 (0.031)

Calibration intercept*

0.00

0.02

0.00

0.04

0.07

Calibration intercept* within clusters ‡

-0.02 (0.442)

-0.00 (0.453)

0.00 (0.000)

0.01 (0.494)

0.05 (0.500)

Calibration slope

1.00

0.97

1.05

0.96

0.92

Calibration slope within clusters ‡

1.05 (0.090)

1.01 (0.092)

1.06 (0.000)

1.00 (0.022)

0.96 (0.021)

  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).