Figure 4

Variation in F-score across 130 simulations. Points beyond the end of the whiskers denote outliers. An F-score of 1.0 is ideal. The LD algorithm was used to select variables for approaches that do not perform variable selection. SNR=0.25 is not shown. S denotes the sparsity setting or the proportion of influential variables in the data. a. β=±1 and b. \({\beta \in \mathcal {U}}\).