Table 2: Performance comparison of different procedures with LOOCV for the colon cancer study.

MethodsNo. of genesLOOCV error rateLOOCV accuracy

Bayesian -prior60.1452 (9/62)0.8548 (53/62)
Bayesian -prior100.1452 (9/62)0.8548 (53/62)
Bayesian -prior140.1129 (7/62)0.8871 (55/62)
SVMa10000.0968 (6/62)0.9032 (56/62)
Classification treeb2000.1452 (9/62)0.8548 (53/62)
1-Nearest-neighborb250.1452 (9/62)0.8548 (53/62)
LogitBoost, estimatedb250.1935 (12/62)0.8065 (50/62)
LogitBoost, 100 iterationsb100.1452 (9/62)0.8548 (53/62)
AdaBoost, 100 iterationsb100.1613 (10/62)0.8387 (52/62)
MAVE-LDc500.1613 (10/62)0.8387 (52/62)
IRWPLSd200.1129 (7/62)0.8871 (55/62)
SGLassoe190.1290 (8/62)0.8710 (54/62)
MRMS + SVM + D1f50.1290 (8/62)0.8710 (54/62)
MRMS + SVM + D2f330.1452 (9/62)0.8548 (53/62)
-test + probit regression60.1452 (9/62)0.8548 (53/62)
-test + probit regression100.1774 (11/62)0.8226 (51/62)
-test + probit regression140.2258 (14/62)0.7742 (48/62)

Proposed by Furey et al. [41].
bProposed by Dettling andBühlmann [42].
cProposed by Antoniadis et al. [43].
dProposed by Ding and Gentleman [44].
eProposed by Ma et al. [45].
fProposed by Maji and Paul [46].