Skip to main content

Table 6 Mean (SD) of calibration slope for testing set over 300 simulations

From: Combined performance of screening and variable selection methods in ultra-high dimensional data in predicting time-to-event outcomes

Selection approach Weak signal Strong signal
n = 2000 n = 4000 n = 2000 n = 4000
ISIS-LASSO 0.345 (0.340) 0.944 (0.065) 0.646 (0.347) 0.965 (0.058)
ISIS-ALASSO 0.318 (0.343) 0.945 (0.065) 0.628 (0.373) 0.965 (0.058)
ISIS-RSF 0.205 (0.379) 0.936 (0.095) 0.486 (0.515) 0.965 (0.059)
SIS 0.210 (0.395) 0.887 (0.126) 0.433 (0.512) 0.955 (0.076)
SIS-LASSO 0.391 (0.336) 0.895 (0.091) 0.624 (0.326) 0.955 (0.076)
SIS-ALASSO 0.321 (0.373) 0.897 (0.091) 0.611 (0.339) 0.955 (0.076)
SIS-RSF 0.092 (2.247) 0.881 (0.129) 0.448 (0.583) 0.950 (0.080)
PSIS 0.004 (0.006) 0.245 (0.055) 0.013 (0.017) 0.370 (0.060)
PSIS-LASSO 0.077 (0.049) 0.355 (0.060) 0.200 (0.072) 0.530 (0.056)
PSIS-ALASSO 0.075 (0.046) 0.376 (0.072) 0.246 (0.116) 0.605 (0.100)
PSIS-RSF 0.078 (0.055) 0.574 (0.135) 0.241 (0.114) 0.806 (0.097)
LASSO 0.359 (0.126) 0.381 (0.076) 0.213 (0.086) 0.219 (0.055)
ALASSO 0.780 (0.131) 0.732 (0.069) 0.642 (0.115) 0.653 (0.069)
  1. SD standard deviation