Modeling Time-Varying Random Objects and Dynamic Networks

TL;DR

This paper introduces a novel framework for analyzing time-varying complex objects like networks using Fréchet means and functional data analysis, enabling insights into their dynamic behavior without relying on algebraic operations.

Contribution

It develops a generalized mean trajectory approach for non-Euclidean data, facilitating functional analysis of dynamic networks and distributions.

Findings

Effective representation of dynamic networks as functional data.

Ability to analyze empirical dynamics such as regression and explosive behavior.

Asymptotic properties of estimators are established.

Abstract

Samples of dynamic or time-varying networks and other random object data such as time-varying probability distributions are increasingly encountered in modern data analysis. Common methods for time-varying data such as functional data analysis are infeasible when observations are time courses of networks or other complex non-Euclidean random objects that are elements of general metric spaces. In such spaces, only pairwise distances between the data objects are available and a strong limitation is that one cannot carry out arithmetic operations due to the lack of an algebraic structure. We combat this complexity by a generalized notion of mean trajectory taking values in the object space. For this, we adopt pointwise Fr\'echet means and then construct pointwise distance trajectories between the individual time courses and the estimated Fr\'echet mean trajectory, thus representing the time-varying objects and networks by functional data. Functional principal component analysis of these distance trajectories can reveal interesting features of dynamic networks and object time courses and is useful for downstream analysis. Our approach also makes it possible to study the empirical dynamics of time-varying objects, including dynamic regression to the mean or explosive behavior over time. We demonstrate desirable asymptotic properties of sample based estimators for suitable population targets under mild assumptions. The utility of the proposed methodology is illustrated with dynamic networks, time-varying distribution data and longitudinal growth data.