HighDimMixedModels.jl: Robust high-dimensional mixed-effects models across omics data
Fig 3
Estimation of β3 (A) and β4 (B) from final model fits in all gene expression simulation settings with q = 3 random effects.
Panel A shows the distributions of estimates of an unpenalized regression coefficient, β3, and panel B shows the distributions of estimates of a penalized coefficient, β4. Numbers at the bottom of the plotting windows in panel B indicate the number of data sets for which the coefficient was incorrectly estimated 0. The box plots represent the distribution of the estimator across all other data sets (i.e. all data sets in which the estimate was non-zero). The true parameter value is represented by a solid black line.