Generalizing the distribution of missing sums in sumsets (Q2672014): Difference between revisions

A natural question in the area of combinatorial number theory is the following: given \(p \in (0, 1)\), construct the set \(A \subseteq \{0,\dots, n-1\}\) by taking each \(i\) in \(A\) independently with probability \(p\). The question is then to study \(|A+A|\). The study of the expectation of \(|A+A|\) was initiated by Martin and O'Bryant, who obtained the expected value for \(p=1/2\). Lazarev, Miller, and O'Bryant [\textit{O. Lazarev} et al., Exp. Math. 22, No. 2, 132--156 (2013; Zbl 1309.11010)] extended this result by studying the variance of the random variable \(|A+A|\) for \(p=1/2\), as well as bounds on the number of missing sums in \(A+A\). This paper continues this investigation by studying the variance for \(p\neq 1/2\), as well as missing sums distribution.