LOGGLE Test

This repository contains all relevant code (functions, simulations, real data applications, etc.) and data used in the paper: Yang, J. & Peng, J. (2018), Estimating Time-Varying Graphical Models, arXiv:1804.03811. Specifically, it includes:

• loggle_test.R: R script for simulation and real data application (method loggle, kernel and invar).

• loggle_test_function.R: Helper functions used in loggle_test.R.

• single_test.R: R script for simulation (method single).

• loggle_1.0.tar.gz: Pre-packaged binary of R package loggle for Linux/Mac OS X.

• loggle_1.0.zip: Pre-packaged binary of R package loggle for Windows.

• SINGLE_1.3.tar.gz: Pre-packaged binary of R package SINGLE.

• loggle_data_100.rda: Data used in simulation of p=100 time-variant graph model.

• loggle_data_500.rda: Data used in simulation of p=500 time-variant graph model.

• loggle_data_100_invar.rda: Data used in simulation of p=100 time-invariant graph model.

• loggle_select_result_real.rda: Pre-stored model selection results in real data application.