Nonuniform Sampling Framework for Granger Causality Analysis

To solve the unreliability issue of Granger causality (GC) inference with uniformly sampled data, we establish a framework of spectrum-based nonparametric GC analysis for nonuniformly sampled time series. Applying this framework to pulse-coupled nonlinear neuronal networks, we demonstrate that, for such nonlinear networks with nonuniformly sampled data, reliable GC inference can be achieved at a low nonuniform mean sampling rate at which the traditional uniform sampling GC may lead to spurious causal inference.