Table 5.

Multivariate Analysis. Relevance was defined as the ratio of the number of non-zero coefficients in the empirical data set to the number of non-zero coefficients above the 95th quantile of the simulation, expressed as a percentage. Absolute value represents the number of non-zero coefficients above the 95th quantile. Alpha indicates the regularisation parameter with the highest relative Relevance. Candidates refer to the number of non-zero coefficients for the model with the specified Alpha. The higher the relevance, the more markers are expected to be proportionally more relevant.

D-dimers				IL-6
Relevance	Absolute	Alpha	Candidates	Relevance	Absolute	Alpha	Candidates
7%	3	0.0143	41	8%	3	0.0229	36
28%	8	0.0464	29	14%	3	0.0744	21
15%	3	0.0744	20	10%	2	0.0588	20