A non-negative spike-and-slab lasso generalized linear stacking prediction modeling method for high-dimensional omics data

Table 1 The parameters of six different simulation scenarios (N = 500, M = 1000)

Scenarios	Non-zero coefficients								Correlation coefficient r	Residual variance \(\sigma\)	^aAdjusted generalized R²
	Group 1			Group 5			Group 20
	β₅	β ₂₀	β ₄₀	β ₂₁₀	β ₂₂₀	β ₂₄₀	β ₉₇₅	β ₉₉₅
1	0.80	− 0.70	1.00	− 0.90	− 0.80	0.90	− 1.00	0.70	0.60	1.60	0.50
2	0.80	− 0.70	1.00	− 0.90	− 0.80	0.90	− 1.00	0.70	0.60	2.60	0.25
3	0.80	− 0.70	1.00	− 0.90	− 0.80	0.90	− 1.00	0.70	0.60	4.50	0.10
4	0.80	− 0.30	1.40	− 0.90	− 0.80	0.90	− 1.50	0.20	0.60	1.80	0.50
5	0.80	− 0.30	1.40	− 0.90	− 0.80	0.90	− 1.50	0.20	0.60	3.10	0.25
6	0.80	− 0.30	1.40	− 0.90	− 0.80	0.90	− 1.50	0.20	0.60	5.50	0.10

The final adjusted generalized R² is adjusted through \(\sigma\)
^aAdjusted generalized R² was obtained by fitting all variables (M = 1000) with the logistic regression model using a large sample (N = 20,000)

ISSN: 1471-2105