Characterization of zero asymptotic variance for component functionals

Let ψ\psi be a graph functional satisfying the paper's growth condition, let pTp_T denote the relevant probability associated with a finite tree TT, and let σψ2\sigma^2_\psi be the asymptotic variance in Theorem~TGpsi. Assume the degree-distribution conditions -- and the supercriticality and positivity conditions --. Zero-variance conjecture.

\sigma^2_\psi=0\iff \psi(T)=0\text{ for every tree $T$ with $|T|>1$ and $p_T>0$.

The trivial examples in Example~Etrivial show that several contributions can force zero variance; the conjecture asserts that, under the stated assumptions, these are the only such cases. The paper explains that values of ψ\psi on K1\mathsf K_1, trees with pT=0p_T=0, and non-trees do not affect σψ2\sigma^2_\psi.

Sources & referencesView supporting material

Primary source

Svante Janson, “Asymptotic normality in random graphs with given vertex degrees”, arXiv:1812.08063 (2019).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.