%I #29 Sep 14 2019 18:01:33
%S 2,7,0,6,7,0,5,6,6,4,7,3,2,2,5,3,8,3,7,8,7,9,9,8,9,8,9,9,4,4,9,6,8,8,
%T 0,6,8,1,5,2,6,3,0,9,1,8,1,9,1,5,1,7,6,2,9,3,6,3,1,7,7,4,5,3,0,8,1,4,
%U 6,7,4,8,2,0,2
%N Decimal expansion of 2/e^2.
%C Decimal expansion of the asymptotic (n -> inf) probability of success in the secretary problem when the number of applicants is uniformly distributed on {1, 2, ..., n}. It means that less is known than in the basic secretary problem, so this constant is less than A068985 (and also A246665).
%H E. L. Presman and I. M. Sonin, <a href="http://mi.mathnet.ru/eng/tvp4342">The best choice problem for a random number of objects</a>, Teor. Veroyatnost. i Primenen., 17:4 (1972), 695-706; <a href="https://doi.org/10.1137/1117078">English version</a>, Theory Probab. Appl., 17:4 (1973), 657-668.
%H Eric Weisstein's MathWorld, <a href="http://mathworld.wolfram.com/SultansDowryProblem.html">Sultan's Dowry Problem</a>
%H Wikipedia, <a href="http://en.wikipedia.org/wiki/Secretary_problem">Secretary problem</a>
%e 0.2706705664732253837879989899...
%t N[2/E^2, 100] // RealDigits // First
%Y Equals twice A092553.
%Y Cf. A068985, A246665.
%K nonn,cons
%O 0,1
%A _José María Grau Ribas_, Sep 08 2019