Skip to main content
. Author manuscript; available in PMC: 2010 Oct 26.
Published in final edited form as: J Am Stat Assoc. 2007 Jun 1;102(478):560–572. doi: 10.1198/016214507000000112

Table 2.

The Goodness-of-Fit, Penalty, and L-Measure Values for Various Models Under First-, Last-, and Hierarchical-Activation Schemes for the Melanoma Dataset

Model Melanoma
First
Last
HA–Bin
Mixture
G P L G P L G P L G P L
1(a) 206.98 421.93 628.91 206.71 463.99 670.70 210.16 438.00 648.16 209.15 419.40 628.55
1(b) 206.29 438.30 644.59 206.29 438.30 644.59 206.29 438.30 644.59 206.29 438.30 644.59
1(c) 207.93 423.87 631.80 206.87 459.21 666.08 210.09 437.96 648.05 207.78 422.63 630.41
1(d) 209.35 409.10 618.45 206.81 492.75 699.56 206.05 436.08 642.13 210.75 405.57 616.32
2 206.59 414.22 620.81 205.43 459.62 665.05 203.50 425.25 628.75 205.65 413.49 619.14

NOTE: The model with the best L-score in each column is highlighted.