Figure 5

Variation in Spearman’s rank correlation ( \(\beta \in \boldsymbol{\mathcal {U}}\) ) across 130 simulations are shown for approaches that do not perform variable selection. Points beyond the end of the whiskers denote outliers. A correlation of 1.0 is ideal. Higher correlation indicates that the approach is able to capture the original variable rank more accurately. SNR=0.25 is not shown. S denotes the sparsity setting or the proportion of influential variables in the data.