This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.
%I A139526 #23 Jun 13 2023 07:16:26 %S A139526 1,1,2,1,6,9,1,12,48,64,1,20,150,500,625,1,30,360,2160,6480,7776,1,42, %T A139526 735,6860,36015,100842,117649,1,56,1344,17920,143360,688128,1835008, %U A139526 2097152,1,72,2268,40824,459270,3306744,14880348,38263752,43046721,1,90,3600,84000,1260000,12600000,84000000,360000000,900000000,1000000000 %N A139526 Triangle A061356 read right to left. %C A139526 Related to the two Appell sequences the Bernoulli polynomials B(n,x) and their umbral compositional inverses (cf. A074909) Up(n,x) = [(x+1)^(n+1)-x^(n+1)] / (n+1). With offset 0, the row polynomials of this entry P(n,x) = (Up(n,0))^(-n) * [x + Up(n,0)]^n = (n+1)^n * [x + 1/(n+1)]^n. Compare to the Abel polynomials of A061356, which are also an Appell sequence. - _Tom Copeland_, Nov 14 2014 %D A139526 R. L. Graham, D. E. Knuth and O. Patashnik, Concrete Mathematics. Addison-Wesley, Reading, MA. Second ed. 1994. %D A139526 Peter D. Schumer (2004), Mathematical Journeys, page 168, Proposition 16.1 (c) %H A139526 P. Bala, <a href="/A251592/a251592.pdf">Fractional iteration of a series inversion operator</a> %H A139526 Wikipedia, <a href="https://en.wikipedia.org/wiki/Lambert_W_function">Lambert W function</a> %F A139526 E.g.f. (with offset 1) Sum_{n >= 1} (1 + n*t)^(n-1)*x^n/n! = x + (1 + 2*t)*x^2/2! + (1 + 6*t + 9*t^2)*x^3/3! + .... For properties of this function see Graham et al., equations 5.60, 5.61 and 7.71. The e.g.f. is the series reversion with respect to x of the function log(1 + x)/(1 + x)^t, which is the e.g.f. for a signed version of A028421. - _Peter Bala_, Jul 18 2013 %F A139526 From _Peter Bala_, Nov 16 2015: (Start) %F A139526 E.g.f. with offset 0 and constant term 1: A(x,t) = ( Sum_{n >= 0} (n + 1)^(n-1)*t^n*x^n/n! )^(1/t). This is the generalized exponential series E_t(x) in the terminology of Graham et al., Section 5.4. %F A139526 A(x,t)^m = 1 + Sum_{n >= 1} m*(m + n*t)^(n-1)*x^n/n!. %F A139526 log(A(x,t)) = Sum_{n >= 1} (n*t)^(n-1)*x^n/n! = 1/t*T(t*x), where T(z) is Euler's tree function. See A000169. %F A139526 A(x,t) = ( 1/x* Revert( x*exp(-x*t)) )^(1/t), where Revert is the series reversion operator with respect to x. %F A139526 In the notation of the Bala link the e.g.f. is I^t(e^x), where I^t is a fractional series inversion operator. Cf. A251592, which has o.g.f. I^t(1 + x), and A260687, which has o.g.f. I^t(1/(1 - x)). (End) %e A139526 (1) times (1) = (1) %e A139526 (1 1) * (1 2) = (1 2) %e A139526 (1 2 1 ) * (1 3 9) = (1 6 9) %e A139526 (1 3 3 1) * (1 4 16 64) = (1 12 48 64) %e A139526 etc. %p A139526 A061356 := proc(n,k) binomial(n-2,k-1)*(n-1)^(n-k-1); end: A139526 := proc(n,k) A061356(n,n-k-1) ; end: for n from 2 to 14 do for k from 0 to n-2 do printf("%d,",A139526(n,k)) ; od: od: # _R. J. Mathar_, May 22 2008 %t A139526 T[n_, k_] := (n - 1)^k*Binomial[n - 2, n - k - 2]; %t A139526 Table[T[n, k], {n, 2, 11}, {k, 0, n - 2}] // Flatten (* _Jean-François Alcover_, Jun 13 2023 *) %o A139526 (PARI) for(n=2,12,forstep(k=n-1,1,-1,print1(binomial(n-2, k-1)*(n-1)^(n-k-1)","))) \\ Herman Jamke (hermanjamke(AT)fastmail.fm), May 10 2008 %Y A139526 Cf. A000272 (row sums), A061356 (row reverse), A028421, A074909, A000169 (main diagonal), A251592, A260687. %K A139526 nonn,tabl %O A139526 2,3 %A A139526 _Alford Arnold_, Apr 24 2008 %E A139526 More terms from _R. J. Mathar_ and Herman Jamke (hermanjamke(AT)fastmail.fm), May 01 2008