A004747 Triangle read by rows: the Bell transform of the triple factorial numbers A008544 without column 0.
1, 2, 1, 10, 6, 1, 80, 52, 12, 1, 880, 600, 160, 20, 1, 12320, 8680, 2520, 380, 30, 1, 209440, 151200, 46480, 7840, 770, 42, 1, 4188800, 3082240, 987840, 179760, 20160, 1400, 56, 1, 96342400, 71998080, 23826880, 4583040, 562800, 45360, 2352, 72, 1
Offset: 1
Examples
Triangle begins: 1; 2, 1; 10, 6, 1; 80, 52, 12, 1; 880, 600, 160, 20, 1; 12320, 8680, 2520, 380, 30, 1; 209440, 151200, 46480, 7840, 770, 42, 1; Tree combinatorics for T(3,2)=6: Consider first the unordered forest of m=2 plane trees with n=3 vertices, namely one vertex with out-degree r=0 (root) and two different trees with two vertices (one root with out-degree r=1 and a leaf with r=0). The 6 increasing labelings come then from the forest with rooted (x) trees x, o-x (1,(3,2)), (2,(3,1)) and (3,(2,1)) and similarly from the second forest x, x-o (1,(2,3)), (2,(1,3)) and (3,(1,2)).
Links
- G. C. Greubel, Rows n = 1..50 of the triangle, flattened
- F. Bergeron, Ph. Flajolet and B. Salvy, Varieties of increasing trees, Lecture Notes in Computer Science vol. 581, ed. J.-C. Raoult, Springer 1992, pp. 24-48.
- P. Blasiak, K. A. Penson and A. I. Solomon, The general boson normal ordering problem, arXiv:quant-ph/0402027, 2004.
- Richell O. Celeste, Roberto B. Corcino, and Ken Joffaniel M. Gonzales. Two Approaches to Normal Order Coefficients, Journal of Integer Sequences, Vol. 20 (2017), Article 17.3.5.
- Tom Copeland, A Class of Differential Operators and the Stirling Numbers
- Milan Janjic, Some classes of numbers and derivatives, JIS 12 (2009) #09.8.3.
- Wolfdieter Lang, On generalizations of Stirling number triangles, J. Integer Seqs., Vol. 3 (2000), #00.2.4.
- Wolfdieter Lang, Combinatorial Interpretation of Generalized Stirling Numbers, J. Int. Seqs. Vol. 12 (2009) #09.3.3.
- Mathias Pétréolle and Alan D. Sokal, Lattice paths and branched continued fractions. II. Multivariate Lah polynomials and Lah symmetric functions, arXiv:1907.02645 [math.CO], 2019.
- Index entries for sequences related to Bessel functions or polynomials
Crossrefs
Programs
-
Magma
function T(n,k) // T = A004747 if k eq 0 then return 0; elif k eq n then return 1; else return (3*(n-1)-k)*T(n-1,k) + T(n-1,k-1); end if; end function; [T(n,k): k in [1..n], n in [1..12]]; // G. C. Greubel, Oct 03 2023
-
Maple
T := (n, m) -> 3^n/m!*(1/3*m*GAMMA(n-1/3)*hypergeom([1-1/3*m, 2/3-1/3*m, 1/3-1/3*m], [2/3, 4/3-n], 1)/GAMMA(2/3)-1/6*m*(m-1)*GAMMA(n-2/3)*hypergeom( [1-1/3*m, 2/3-1/3*m, 4/3-1/3*m], [4/3, 5/3-n], 1)/Pi*3^(1/2)*GAMMA(2/3)): for n from 1 to 6 do seq(simplify(T(n,k)),k=1..n) od; # Karol A. Penson, Feb 06 2004 # The function BellMatrix is defined in A264428. # Adds (1,0,0,0, ..) as column 0. BellMatrix(n -> mul(3*k+2, k=(0..n-1)), 9); # Peter Luschny, Jan 29 2016
-
Mathematica
(* First program *) T[1,1]= 1; T[, 0]= 0; T[0, ]= 0; T[n_, m_]:= (3*(n-1)-m)*T[n-1, m]+T[n-1, m-1]; Flatten[Table[T[n, m], {n,12}, {m,n}] ][[1 ;; 45]] (* Jean-François Alcover, Jun 16 2011, after recurrence *) (* Second program *) f[n_, m_]:= m/n Sum[Binomial[k, n-m-k] 3^k (-1)^(n-m-k) Binomial[n+k-1, n-1], {k, 0, n-m}]; Table[n! f[n, m]/(m! 3^(n-m)), {n,12}, {m,n}]//Flatten (* Michael De Vlieger, Dec 23 2015 *) (* Third program *) rows = 12; T[n_, m_]:= BellY[n, m, Table[Product[3k+2, {k, 0, j-1}], {j, 0, rows}]]; Table[T[n, m], {n,rows}, {m,n}]//Flatten (* Jean-François Alcover, Jun 22 2018 *)
-
Sage
# uses [bell_transform from A264428] triplefactorial = lambda n: prod(3*k+2 for k in (0..n-1)) def A004747_row(n): trifact = [triplefactorial(k) for k in (0..n)] return bell_transform(n, trifact) [A004747_row(n) for n in (0..10)] # Peter Luschny, Dec 21 2015
Formula
T(n, m) = n!*A048966(n, m)/(m!*3^(n-m));
T(n+1, m) = (3*n-m)*T(n, m)+ T(n, m-1), for n >= m >= 1, with T(n, m) = 0, for n
E.g.f. of m-th column: ( 1 - (1-3*x)^(1/3) )^m/m!.
Sum_{k=1..n} T(n, k) = A015735(n).
For a formula expressed as special values of hypergeometric functions 3F2 see the Maple program below. - Karol A. Penson, Feb 06 2004
T(n,1) = A008544(n-1). - Peter Luschny, Dec 23 2015
Extensions
New name from Peter Luschny, Dec 21 2015
A051141 Triangle read by rows: a(n, m) = S1(n, m)*3^(n-m), where S1 are the signed Stirling numbers of first kind A008275 (n >= 1, 1 <= m <= n).
1, -3, 1, 18, -9, 1, -162, 99, -18, 1, 1944, -1350, 315, -30, 1, -29160, 22194, -6075, 765, -45, 1, 524880, -428652, 131544, -19845, 1575, -63, 1, -11022480, 9526572, -3191076, 548289, -52920, 2898, -84, 1, 264539520, -239660208
Offset: 1
Comments
Previous name was: Generalized Stirling number triangle of first kind.
a(n,m) = R_n^m(a=0,b=3) in the notation of the given reference.
a(n,m) is a Jabotinsky matrix, i.e., the monic row polynomials E(n,x) := Sum_{m=1..n} a(n,m)*x^m = Product_{j=0..n-1} (x - 3*j), n >= 1 and E(0,x) := 1 are exponential convolution polynomials (see A039692 for the definition and a Knuth reference).
This is the signed Stirling1 triangle with diagonals d>=0 (main diagonal d=0) scaled with 3^d.
Exponential Riordan array [1/(1 + 3*x), log(1 + 3*x)/3]. The unsigned triangle is [1/(1 - 3*x), log(1/(1 - 3*x)^(1/3))]. - Paul Barry, Apr 29 2009
Also the Bell transform of the triple factorial numbers A032031 which adds a first column (1, 0, 0 ...) on the left side of the triangle and computes the unsigned values. For the definition of the Bell transform, see A264428. See A004747 for the triple factorial numbers A008544 and A203412 for the triple factorial numbers A007559 as well as A039683 and A132062 for the case of double factorial numbers. - Peter Luschny, Dec 21 2015
Examples
Triangle starts: 1; -3, 1; 18, -9, 1; -162, 99, -18, 1; 1944, -1350, 315, -30, 1; -29160, 22194, -6075, 765, -45, 1; 524880, -428652, 131544, -19845, 1575, -63, 1; --- Row polynomial E(3,x) = 18*x-9*x^2+x^3. From _Paul Barry_, Apr 29 2009: (Start) The unsigned array [1/(1 - 3*x), log(1/(1 - 3*x)^(1/3))] has production matrix 3, 1; 9, 6, 1; 27, 27, 9, 1; 81, 108, 54, 12, 1; 243, 405, 270, 90, 15, 1; 729, 1458, 1215, 540, 135, 18, 1; ... which is A007318^{3} beheaded (by viewing A007318 as a lower triangular matrix). See the comment above. (End)
Links
- G. C. Greubel, Table of n, a(n) for the first 50 rows, flattened
- Richell O. Celeste, Roberto B. Corcino, and Ken Joffaniel M. Gonzales, Two Approaches to Normal Order Coefficients, Journal of Integer Sequences, Vol. 20 (2017), Article 17.3.5.
- Wolfdieter Lang, First 10 rows.
- D. S. Mitrinovic and M. S. Mitrinovic, Tableaux d'une classe de nombres reliés aux nombres de Stirling, Univ. Beograd. Publ. Elektrotehn. Fak. Ser. Mat. Fiz. 77 (1962), 1-77.
Crossrefs
Programs
-
Mathematica
a[n_, m_] /; n >= m >= 1 := a[n, m] = a[n-1, m-1] - 3(n-1)*a[n-1, m]; a[n_, m_] /; n < m = 0; a[, 0] = 0; a[1, 1] = 1; Flatten[Table[a[n, m], {n, 1, 9}, {m, 1, n}]][[1 ;; 38]] (* _Jean-François Alcover, Jun 01 2011, after formula *) Table[StirlingS1[n, m]*3^(n - m), {n, 1, 10}, {m, 1, n}]//Flatten (* G. C. Greubel, Oct 24 2017 *)
-
PARI
for(n=1,10, for(m=1,n, print1(stirling(n,m,1)*3^(n-m), ", "))) \\ G. C. Greubel, Oct 24 2017
-
Sage
# uses[bell_transform from A264428] triplefactorial = lambda n: 3^n*factorial(n) def A051141_row(n): trifact = [triplefactorial(k) for k in (0..n)] return bell_transform(n, trifact) [A051141_row(n) for n in (0..8)] # Peter Luschny, Dec 21 2015
Formula
a(n, m) = a(n-1, m-1) - 3*(n-1)*a(n-1, m) for n >= m >= 1; a(n, m) = 0 for n < m; a(n, 0) = 0 for n >= 1; a(0, 0) = 1.
E.g.f. for the m-th column of the signed triangle: (log(1 + 3*x)/3)^m/m!.
|a(n,1)| = A032031(n-1). - Peter Luschny, Dec 23 2015
Extensions
Name clarified using a formula of the author by Peter Luschny, Dec 23 2015
A265606 Triangle read by rows: The Bell transform of the quartic factorial numbers (A007696).
1, 0, 1, 0, 1, 1, 0, 5, 3, 1, 0, 45, 23, 6, 1, 0, 585, 275, 65, 10, 1, 0, 9945, 4435, 990, 145, 15, 1, 0, 208845, 89775, 19285, 2730, 280, 21, 1, 0, 5221125, 2183895, 456190, 62965, 6370, 490, 28, 1, 0, 151412625, 62002395, 12676265, 1715490, 171255, 13230, 798, 36, 1
Offset: 0
Examples
[1], [0, 1], [0, 1, 1], [0, 5, 3, 1], [0, 45, 23, 6, 1], [0, 585, 275, 65, 10, 1], [0, 9945, 4435, 990, 145, 15, 1], [0, 208845, 89775, 19285, 2730, 280, 21, 1],
Links
- Richell O. Celeste, Roberto B. Corcino, and Ken Joffaniel M. Gonzales. Two Approaches to Normal Order Coefficients, Journal of Integer Sequences, Vol. 20 (2017), Article 17.3.5.
- Peter Luschny, The Bell transform
Crossrefs
Programs
-
Mathematica
(* The function BellMatrix is defined in A264428. *) rows = 10; M = BellMatrix[Pochhammer[1/4, #] 4^# &, rows]; Table[M[[n, k]], {n, 1, rows}, {k, 1, n}] // Flatten (* Jean-François Alcover, Jul 23 2019 *)
-
Sage
# uses[bell_transform from A264428] def A265606_row(n): multifact_4_1 = lambda n: prod(4*k + 1 for k in (0..n-1)) mfact = [multifact_4_1(k) for k in (0..n)] return bell_transform(n, mfact) [A265606_row(n) for n in (0..7)]
Comments