Biases in prime factorizations and Liouville functions for arithmetic progressions (Q2199486)

Let \(\lambda(n)\) be the Liouville function, that is defined as \(\lambda(n) = (-1)^{\Omega(n)}\) where \(\Omega(n)\) is the total number of prime factors of the positive integer \(n\) (counting its multiplicity). It is well known that the Riemann hypothesis is equivalent to the statement that \(L(x) := \sum_{n \le x} \lambda(n) = O_\varepsilon(x^{1/2+\varepsilon})\) for any \(\varepsilon > 0\), whereas the prime number theorem is equivalent to the estimate \(o(x)\). This paper introduces some refinements of the Liouville function which detect how primes in given arithmetic progressions appear in prime factorizations, and studies its average order. With this aim, let \(\Omega(n; q, a)\) be the number of prime factors of \(n\) congruent to \(a\) modulo \(q\), and \(\lambda(n; q, a) = (-1)^{\Omega(n; q, a)}\); then, a natural question is to ask for the asymptotic behavior of \(L(x; q, a) := \sum_{n \le x} \lambda(n; q, a)\). Yet more, the paper defines the \(r\)-fold products \[ \lambda(n ; q, a_{1}, \dots, a_{r}) = \prod_{i=1}^{r} \lambda(n ; q, a_{i}) \] where the \(a_i\) are distinct residue classes modulo \(q\), with \(1 \le r \le q\), and define \(\Omega(n;q,a_1,\dots,a_r)\) and \(L(x;q,a_1,\dots,a_r)\) analogously, and studies the order or \(L(x;q,a_1,\dots,a_r)\). For instance, it is shown that, for \(q \ge 2\) and \(a_1,\dots,a_{\varphi(q)}\) the residue classes modulo \(q\) such that \(\gcd(a_i,q) = 1\), we have \[ \sum_{n \leq x} \lambda(n ; q, a_{1}, \dots, a_{\varphi(q)}) = o(x) \] (and \(O_\varepsilon(x^{1/2+\varepsilon}\) if we assume the Riemann hypothesis). However, if \(b_1,\dots,b_{q-\varphi(q)}\) are the remaining residues classes modulo \(q\), and we take any subset of size \(k\), the asymptotic behavior of \(\sum_{n \leq x} \lambda(n ; q, b_{1}, \dots, b_{k})\) is different of what happens with \(a_{1}, \dots, a_{\varphi(q)}\). Moreover, supported by numerical tests, the paper analyzes analogues of Pólya's conjecture (i.e., that \(L(x) \le 0\) for \(x \ge 2\), that where disproved by Haselgrove in 1958), and proves results related to the sign changes of the associated summatory functions.