The C1C2: A framework for simultaneous model selection and assessment

BMC Bioinformatics

Table 1 Coefficient estimates of model (6) with w_{i 1}= ${‖ \bar{\hat{Δ}} - Δ ‖}_{i}$ , i = 1, ..., 128, as a response variable.

	Estimate	Std.Error	t-value	Pr(>\|t\|)
intercept	-0.01992	0.04606	-0.433	0.6661
c1c2	-0.04337	0.03761	-1.153	0.2511
ga	0.15683	0.03761	4.170	5.72e-05
cor	0.07211	0.03761	1.918	0.0575
15	0.21324	0.03761	5.670	9.75e-08
all	0.32754	0.03761	8.710	1.78e-14

c1c2 – the C¹C² was used (as opposed to repeated K-fold cross-validation), ga – the GA search method was used (as opposed to the brute force search method), cor – correlated independent variables in the dataset (as opposed to uncorrelated,) 15 – n = 15 observations in the dataset (as opposed to n = 200), all – no assumption regarding the number of nonzero δ_i(as opposed to the assumption of three δ_i= 1).

ISSN: 1471-2105