sunpeng52/ggm

R codes for "Sparse positive-definite estimation for covariance matrices with repeated measurements"

GGM

R codes for "Sparse positive-definite estimation for covariance matrices with repeated measurements"

Basic Functions

functions.R & functions_2.R: functions to set up covariance matrices, run ADMM, etc.

Figure 1 & Figure 2

mm_generate_data.R: generate simulation data for Model 1 and Model 2 with p =100, 200.

mm_simulation_100.R & mm_simulation_200.R: apply our proposed procedures to the simulated data.

mm_rate_100.R: compute fpr and tpr for Figure 2.

00 mm_analysis_100.R: generate Figure 1.

00 mm_roc_plot.R: generate Figure 2.

Figure 3

mm_comparison_generate_data.R: generate simulation data for Model 3 and Model 4.

mm_comparison_simulation.R: apply our proposed procedures to the simulated data.

mm_comparison_analysis.R: compute several metrics for Figure 3.

00 mm_comparison_plot.R: generate Figure 3.

Table 1 in the Supplementary materials

mm_comparison_soft_admm.R: simulation for comparison between ADMM and soft thresholding, results are summarized in Table 1 in the supplementary materials.

Real Data Analysis

01_Real_Analysis_New.R: code for real data analtsis, generating Figure 4.