Corpus ID: 236428582

A principled (and practical) test for network comparison

  title={A principled (and practical) test for network comparison},
  author={Gecia Bravo Hermsdorff and Lee M. Gunderson and Pierre-Andr{\'e} G. Maugis and Carey E. Priebe},
How might one test the hypothesis that graphs were sampled from the same distribution? Here, we compare two statistical tests that address this question. The first uses the observed subgraph densities themselves as estimates of those of the underlying distribution. The second test uses a new approach that converts these subgraph densities into estimates of the graph cumulants of the distribution. We demonstrate — via theory, simulation, and application to real data — the superior statistical… Expand

Figures from this paper


Practical methods for graph two-sample testing
This paper considers the problem of two-sample testing of large graphs and demonstrates the practical merits and limitations of existing theoretical tests and their bootstrapped variants, and proposes two new tests based on asymptotic distributions that are computationally less expensive and more reliable than the existing methods. Expand
A spectral-based framework for hypothesis testing in populations of networks.
A test to determine whether two given samples of networks come from the same random model or distribution is developed, based on the trace of the third order for a centered and scaled adjacency matrix, which it is proved converges to the standard normal distribution as the number of nodes tends to infinity. Expand
Testing for Equivalence of Network Distribution Using Subgraph Counts
Simulation experiments and an illustrative example on a sample of brain networks where it is found that highly creative individuals’ brains present significantly more short cycles than found in less creative people are presented. Expand
Two-sample hypothesis testing for inhomogeneous random graphs
The study of networks leads to a wide range of high dimensional inference problems. In many practical applications, one needs to draw inference from one or few large sparse networks. The presentExpand
A Semiparametric Two-Sample Hypothesis Testing Problem for Random Graphs
ABSTRACT Two-sample hypothesis testing for random graphs arises naturally in neuroscience, social networks, and machine learning. In this article, we consider a semiparametric problem of two-sampleExpand
Network motifs come in sets: correlations in the randomization process.
  • Reid Ginoza, A. Mugler
  • Mathematics, Medicine
  • Physical review. E, Statistical, nonlinear, and soft matter physics
  • 2010
The most common algorithms used to generate the ensemble from the real network change subgraph counts in a highly correlated manner, such that one subgraph's status as a motif may not be independent from the statuses of the other subgraphs. Expand
Comparing methods for comparing networks
This work reviews and classify a collection of network comparison methods, highlighting the criteria they are based on and their advantages and drawbacks, and applies the methods to two real-world datasets, the European Air Transportation Network and the FAO Trade Network, to discuss the results that can be drawn from this type of analysis. Expand
Geometric Network Comparison
Network analysis has a crucial need for tools to compare networks and assess the significance of differences between networks. We propose a principled statistical approach to network comparison thatExpand
Subsampling large graphs and invariance in networks
Specify a randomized algorithm that, given a very large graph or network, extracts a random subgraph. What can we learn about the input graph from a single subsample? We derive laws of large numbersExpand
Fitting ERGMs on big networks.
  • Weihua An
  • Computer Science, Medicine
  • Social science research
  • 2016
A framework and a series of methods (based on existent algorithms) to address conceptual and computational issues for fitting ERGMs on big networks are described. Expand