The Invariance Partial Pruning Approach to The Network Comparison in Longitudinal Data
Existing methods to compare idiographic networks are global tests that indicate the presence/absence of difference. They typically release the equality constrains for all edges once the global test indicates heterogeneity, which leads to false positive for group difference in network structures. We therefore present the invariance partial pruning (IVPP) approach, which first evaluate the presence of heterogeneity with network invariance test and then determine the exact locus of edge equality and difference with partial pruning. Simulation results indicated that the invariance test based on AIC and BIC performed better than the invariance test based on LRT, and partial pruning successfully uncovered specific edge difference with high sensitivity and specificity. IVPP is an essential supplement to the existing network methodology by allowing the comparison of networks from time-series and panel data, and also allowing the test of specific edge difference. We implemented the algorithm in the R-package IVPP