Cauchy means of Dirichlet polynomials (Q5965179)

The author proves the following theorem: There exist two positive constants \(c\) and \(C\) such that for all positive integers \(N\), \(k\) and \(\sigma\), \(0\leq\sigma<1\), we have \[ \bigg|\int_{\mathbb R}\Big|\sum_{n=1}^N\frac{1}{n^{\sigma+it/\sqrt{c_{\sigma, N^{k}}}}}\Big|^{2k}\frac{dt}{\pi(t^{2}+1)}-c\Big(\sum_{n=1}^N\frac{1}{n^{\sigma}}\Big)^{2k}\bigg|\leq C\frac{(1-\sigma)\log N}{k^{1/2}}\Big(\sum_{n=1}^{N}\frac{1}{n^{\sigma}}\Big)^{2k}, \] where \[ c_{\sigma, N}=\frac{2}{(1-\sigma)^2}+\mathcal{O}\big(N^{\sigma-1}(\log N)^{2}\big). \] For the proof of this main result the author applies a random model for Dirichlet polynomials proving the following result. Lemma. Let \(0<\sigma<1\). Then \[ \mathbb{E}(\widetilde{Y}^{2})=2\Big\{\sum_{m=1}^{N}\frac{(\log m)^{2}}{m^{\sigma}L_{N}}-(\sum_{m=1}^{N}\frac{(\log m)}{m^{\sigma}L_{N}})^{2}\Big\}=\frac{2}{(1-\sigma)^{2}}+\mathcal{O}\big(N^{\sigma -1}(\log N)^{2}\big), \] where \(L_{n}=\sum_{n=1}^{N}\frac{1}{n^{\sigma}}\) and \(Y\) is a random variable defined by the equation \(\mathbb{P}(Y=\log n)= \frac{1}{n^{\sigma} L^{N}}\).