Skip to main content

Table 2 Model selection frequency for the different measures of goodness-of-fit.

From: An evaluation of R2 as an inadequate measure for nonlinear models in pharmacological and biochemical research: a Monte Carlo approach

  Model R 2 adj AICc BIC resVar red. Chi 2
s.d. = 0.02 L5 228 40 41 231 227
  L4 341 170 154 347 261
  L3 867 1648 1660 857 567
  B5 305 90 91 304 689
  B4      
  B3      
  W4      
  W3      
  baro5 259 52 54 261 256
L3 [%]   43.4 82.4 83.0 42.9 28.4
s.d. = 0.1 L5 124 10 10 122 118
  L4 349 161 145 365 288
  L3 958 1668 1684 939 576
  B5 279 30 30 283 308
  B4 35 53 51 34 361
  B3 5 43 44 6 144
  W4      
  W3      
  baro5 250 35 36 251 205
L3 [%]   47.9 83.4 84.2 47.0 28.8
s.d. = 0.4 L5 42 5 6 43 69
  L4 234 72 65 243 187
  L3 698 982 989 667 374
  B5 55 3 3 61 67
  B4 257 81 76 194 208
  B3 303 575 582 406 335
  W4 51 46 43 89 305
  W3 221 226 226 145 357
  baro5 139 10 10 152 98
L3 [%]   34.9 49.1 49.5 33.4 18.7
  1. Same setup as in Table 1 but summarizing the model selection for each iteration and each measure by selection frequency. A percentage of selecting the 'true' model L3 is given as a summary.