Mazur–Rubin's variance asymptotic conjecture for modular symbols

Let fS2(Γ0(q))f\in S_2(\Gamma_0(q)), where qq is a positive integer. For each positive integer cc, define

Var(f,c)=1ϕ(c)amodc(a,c)=1(a/cE(f,c))2,\operatorname{Var}(f,c)=\frac{1}{\phi(c)}\sum_{\substack{a\bmod c\\(a,c)=1}}\left(\langle a/c\rangle-\operatorname{E}(f,c)\right)^2,

where r=irRe(f(z)dz)\langle r\rangle=\int_{i\infty}^r\operatorname{Re}(f(z)\,dz) and E(f,c)\operatorname{E}(f,c) is the corresponding mean. Mazur–Rubin's variance asymptotic conjecture. There exists a constant CfC_f and, for each divisor dd of qq, a constant Df,dD_{f,d} such that

limc(c,q)=d(Var(f,c)Cflogc)=Df,d.\lim_{\substack{c\to\infty\\(c,q)=d}}\left(\operatorname{Var}(f,c)-C_f\log c\right)=D_{f,d}.

The constant CfC_f is the variance slope and Df,dD_{f,d} is the variance shift. The conjecture predicts the asymptotic growth of the second moment of modular symbols along each fixed gcd class with the level; the supplied text gives no resolution status.

Sources & referencesView supporting material

Primary source

Petru Constantinescu, “Distribution of modular symbols in H^3”, arXiv:2005.07629 (2020).

Additional references

2 papers in this index state this conjecture (2017–2020). The statement above is taken from the most recent of them; the others are arXiv:1703.09526.

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.