12 problems

Conjecture on coefficient-nullity tests from simultaneous motif constraints in ERGMs
arXiv@arxiv
Motif-constraint testing conjecture. Formulating null hypotheses that constrain several motif counts simultaneously could yield tests for coefficient nullity in ERGMs.
Conjectureopenexponential-random-graphs network-models statistical-testing
0
Delocalization conjecture for eigenvectors of general binary networks
arXiv@arxiv
Let $\mathbf{A}=\mathbf{P}+\mathbf{E}$ be the observed adjacency matrix, with rank parameter $r$ , and suppose that Assumptions (deloc)–(moments) hold, except that the moment assump…
Conjectureopeneigenvector-delocalization network-models random-matrix-theory
1
Conjecture on the need for a new estimator of the probability matrix in DCMMs
arXiv@arxiv
Consider the degree-corrected mixed membership model with probability matrix $\boldsymbol{P}$ , membership-related matrix $\boldsymbol{G}$ , and first singular vector…
Conjectureopenmixed-membership-models network-models statistical-estimation
1
Conjecture on additional noise for low-degree nodes in degree-corrected mixed membership models
arXiv@arxiv
In a degree-corrected mixed membership model, let $\theta_i$ denote the degree parameter of node $i$ , let $\boldsymbol{\Theta}$ denote the degree-parameter vector, let…
Conjectureopenmixed-membership-models network-models statistical-estimation
0
Conjecture on graphon methods for asymptotics of multiview latent network models
arXiv@arxiv
Graphon asymptotics conjecture. Studying this asymptotic problem will require approaching it using graphon theory and the limits of dense graph sequences.
Conjectureopenasymptotic-statistics graphons network-models
1
Conjecture on GLAMLE for mixed discrete and continuous network distributions
arXiv@arxiv
Mixed-distribution conjecture. The proposed method should also be able to handle mixtures of discrete and continuous distributions, for example when some $Y_{ij}$ are Bernoulli, so…
Conjectureopenlatent-variable-models network-models statistical-inference
0
Conjecture on edge probabilities depending on socioeconomic and other covariates
arXiv@arxiv
Covariate-dependence conjecture. Edge probabilities should depend on such variables.
Conjectureopeneconomic-networks network-models statistical-inference
0
Extension of SIS persistence asymptotics to Erlang infectious periods
arXiv@arxiv
Let theorem1(i) and theorem2(i) denote the asymptotic results for the persistence time of SIS infections in the model with heterogeneous populations or networks, established under…
Conjectureopenasymptotic-analysis network-models stochastic-epidemics
1
Conjecture on the asymptotic number of estimable parameters in the 2K model
arXiv@arxiv
Asymptotic estimability conjecture. For a single observation, asymptotically at most half of the parameters are estimable. More specifically, the graph $T_n$ has $(n-1/2)^2$ non-ze…
Conjectureopennetwork-models statistical-estimation
1
Consistency of variational approximation for the multi-graph stochastic block model
arXiv@arxiv
The multi-graph stochastic block model consists of multiple network layers sharing node class assignments, with layer-specific connection probability matrices; variational approxim…
Conjectureopennetwork-models stochastic-block-models variational-inference
0
De-biased moment estimator conjecture for sparse network models
arXiv@arxiv
De-biasing conjecture. In this range, an appropriate de-biasing of $b[?mb[0m$ may yield a $\sqrt{n}$ -consistent estimate of the target moment.
Conjectureopenasymptotic-statistics moment-methods network-models
0
Conjecture on the pair-dependent optimal convex-combination parameter
arXiv@arxiv
Let $v_k$ and $v_\ell$ be a pair of vertices in the spatial preferred attachment model, and let $c$ be the parameter in the estimator based on the convex combination $(1-c)L+cU$ of…
Conjectureopengeometric-graphs network-models spatial-preferential-attachment
0