|
%I #19 Nov 12 2014 23:19:05
%S 0,0,0,1,0,1,2,1,2,3,2,5,6,5,6,7,6,5,10,9,8,7,6,7,10,9,10,13,12,13,16,
%T 15,14,15,14,13,16,15,14,15,14,13,16,15,16,17,16,17,16,15,16,17,16,17,
%U 18,17,20,21,20,23,28,27,26,27,26,25,30,29,28,27,26,25,28
%N After A084937(n) has been computed, let m = largest term so far in A084937. Then a(n) = number of positive integers < m that are missing from A084937 at this point.
%C Running count of missing numbers in A084937.
%C It appears that at any point, the number of missing even numbers from A084937 is always much larger than the number of missing odd numbers. It would be nice to have a more precise statement of this property.
%C In this regard, it would be helpful to have two further sequences, one giving the number of even missing numbers at each point, the other giving the number of odd missing numbers. These are now A250099, A250100. See also A249777, A249856, A249867.
%H N. J. A. Sloane, <a href="/A249686/b249686.txt">Table of n, a(n) for n = 1..100000</a>
%e After step 7 of A084937, here is what we have:
%e 1 2 3 4 5 6 7 ... n
%e 1 2 3 5 4 7 9 ... A084937(n)
%e so m = 9, and the missing numbers < 9 are 6 and 8, so a(7) = 2.
%Y Cf. A084937, A250099, A250100. See A249777, A249856, A249857, A249858 for another way of looking at this question.
%K nonn
%O 1,7
%A _N. J. A. Sloane_, Nov 05 2014
|
