Pascal matrix

In matrix theory and combinatorics, a Pascal matrix is a matrix (possibly infinite) containing the binomial coefficients as its elements. It is thus an encoding of Pascal's triangle in matrix form. There are three natural ways to achieve this: as a lower-triangular matrix, an upper-triangular matrix, or a symmetric matrix. For example, the 5 × 5 matrices are: L_5 = \begin{pmatrix} 1 & 0 & 0 & 0 & 0 \\ 1 & 1 & 0 & 0 & 0 \\ 1 & 2 & 1 & 0 & 0 \\ 1 & 3 & 3 & 1 & 0 \\ 1 & 4 & 6 & 4 & 1 \end{pmatrix}\,\,\, U_5 = \begin{pmatrix} 1 & 1 & 1 & 1 & 1 \\ 0 & 1 & 2 & 3 & 4 \\ 0 & 0 & 1 & 3 & 6 \\ 0 & 0 & 0 & 1 & 4 \\ 0 & 0 & 0 & 0 & 1 \end{pmatrix}\,\,\, S_5 = \begin{pmatrix} 1 & 1 & 1 & 1 & 1 \\ 1 & 2 & 3 & 4 & 5 \\ 1 & 3 & 6 & 10 & 15 \\ 1 & 4 & 10 & 20 & 35 \\ 1 & 5 & 15 & 35 & 70 \end{pmatrix}=L_5 \times U_5 There are other ways in which Pascal's triangle can be put into matrix form, but these are not easily extended to infinity.

In matrix theory and combinatorics, a Pascal matrix is a matrix (possibly infinite) containing the binomial coefficients as its elements. It is thus an encoding of Pascal's triangle in matrix form. There are three natural ways to achieve this: as a lower-triangular matrix, an upper-triangular matrix, or a symmetric matrix. For example, the 5 × 5 matrices are: L_5 = \begin{pmatrix} 1 & 0 & 0 & 0 & 0 \\ 1 & 1 & 0 & 0 & 0 \\ 1 & 2 & 1 & 0 & 0 \\ 1 & 3 & 3 & 1 & 0 \\ 1 & 4 & 6 & 4 & 1 \end{pmatrix}\,\,\, U_5 = \begin{pmatrix} 1 & 1 & 1 & 1 & 1 \\ 0 & 1 & 2 & 3 & 4 \\ 0 & 0 & 1 & 3 & 6 \\ 0 & 0 & 0 & 1 & 4 \\ 0 & 0 & 0 & 0 & 1 \end{pmatrix}\,\,\, S_5 = \begin{pmatrix} 1 & 1 & 1 & 1 & 1 \\ 1 & 2 & 3 & 4 & 5 \\ 1 & 3 & 6 & 10 & 15 \\ 1 & 4 & 10 & 20 & 35 \\ 1 & 5 & 15 & 35 & 70 \end{pmatrix}=L_5 \times U_5 There are other ways in which Pascal's triangle can be put into matrix form, but these are not easily extended to infinity. L_{ij} = {i \choose j} = \frac{i!}{j!(i-j)!}, j \le i U_{ij} = {j \choose i} = \frac{j!}{i!(j-i)!}, i \le j S_{ij} = {i+j \choose i} = {i+j \choose j} = \frac{(i+j)!}{i!j!} such that the indices i, j start at 0, and ! denotes the factorial. These matrices along with related variants arise naturally in the study of Hermite and Laguerre polynomials . The matrices have the pleasing relationship S n = L n U n . From this it is easily seen that all three matrices have determinant 1, as the determinant of a triangular matrix is simply the product of its diagonal elements, which are all 1 for both L n and U n . In other words, matrices S n , L n , and U n are unimodular, with L n and U n having trace n. The trace of S n is given by \text{tr}(S_n) = \sum^n_{i=1} \frac{ [ 2(i-1) ] !}{[(i-1)!]^2} = \sum^{n-1}_{k=0} \frac{ (2k) !}{(k!)^2} with the first few terms given by the sequence 1, 3, 9, 29, 99, 351, 1275, ... . The matrix S n is similar to its inverse, and consequently the eigenvalues of S n come in reciprocal pairs, ie. λ is an eigenvalue if and only if 1/λ is an eigenvalue . Consequently, for odd values of n the matrix S n has an eigenvalue of 1. If n = p k for an odd prime p, an eigenvector with eigenvalue 1 is directly related to counting points on elliptic curves over finite fields and local zeta functions . ==Construction== A Pascal matrix can actually be constructed by taking the matrix exponential of a special subdiagonal or superdiagonal matrix . The example below constructs a 7 × 7 Pascal matrix, but the method works for any desired n × n Pascal matrices. The dots in the following matrices represent zero elements. \begin{array}{lll} & L_7=\exp \left ( \left [ \begin{smallmatrix} . & . & . & . & . & . & . \\ 1 & . & . & . & . & . & . \\ . & 2 & . & . & . & . & . \\ . & . & 3 & . & . & . & . \\ . & . & . & 4 & . & . & . \\ . & . & . & . & 5 & . & . \\ . & . & . & . & . & 6 & . \end{smallmatrix} \right ] \right ) = \left [ \begin{smallmatrix} 1 & . & . & . & . & . & . \\ 1 & 1 & . & . & . & . & . \\ 1 & 2 & 1 & . & . & . & . \\ 1 & 3 & 3 & 1 & . & . & . \\ 1 & 4 & 6 & 4 & 1 & . & . \\ 1 & 5 & 10 & 10 & 5 & 1 & . \\ 1 & 6 & 15 & 20 & 15 & 6 & 1 \end{smallmatrix} \right ] \quad \\ \\ & U_7=\exp \left ( \left [ \begin{smallmatrix} {\color{white}1}. & 1 & . & . & . & . & . \\ {\color{white}1}. & . & 2 & . & . & . & . \\ {\color{white}1}. & . & . & 3 & . & . & . \\ {\color{white}1}. & . & . & . & 4 & . & . \\ {\color{white}1}. & . & . & . & . & 5 & . \\ {\color{white}1}. & . & . & . & . & . & 6 \\ {\color{white}1}. & . & . & . & . & . & . \end{smallmatrix} \right ] \right ) = \left [ \begin{smallmatrix} 1 & 1 & 1 & 1 & 1 & 1 & 1 \\ . & 1 & 2 & 3 & 4 & 5 & 6 \\ . & . & 1 & 3 & 6 & 10 & 15 \\ . & . & . & 1 & 4 & 10 & 20 \\ . & . & . & . & 1 & 5 & 15 \\ . & . & . & . & . & 1 & 6 \\ . & . & . & . & . & . & 1 \end{smallmatrix} \right ] \\ \\ \therefore & S_7 =\exp \left ( \left [ \begin{smallmatrix} . & . & . & . & . & . & . \\ 1 & . & . & . & . & . & . \\ . & 2 & . & . & . & . & . \\ . & . & 3 & . & . & . & . \\ . & . & . & 4 & . & . & . \\ . & . & . & . & 5 & . & . \\ . & . & . & . & . & 6 & . \end{smallmatrix} \right ] \right ) \exp \left ( \left [ \begin{smallmatrix} {\color{white}i}. & 1 & . & . & . & . & . \\ {\color{white}i}. & . & 2 & . & . & . & . \\ {\color{white}i}. & . & . & 3 & . & . & . \\ {\color{white}i}. & . & . & . & 4 & . & . \\ {\color{white}i}. & . & . & . & . & 5 & . \\ {\color{white}i}. & . & . & . & . & . & 6 \\ {\color{white}i}. & . & . & . & . & . & . \end{smallmatrix} \right ] \right ) = \left [ \begin{smallmatrix} 1 & 1 & 1 & 1 & 1 & 1 & 1 \\ 1 & 2 & 3 & 4 & 5 & 6 & 7 \\ 1 & 3 & 6 & 10 & 15 & 21 & 28 \\ 1 & 4 & 10 & 20 & 35 & 56 & 84 \\ 1 & 5 & 15 & 35 & 70 & 126 & 210 \\ 1 & 6 & 21 & 56 & 126 & 252 & 462 \\ 1 & 7 & 28 & 84 & 210 & 462 & 924 \end{smallmatrix} \right ]. \end{array} One cannot simply assume exp(A) exp(B) = exp(A + B), for n × n matrices A and B; this equality is only true when AB = BA (i.e. when the matrices A and B commute). In the construction of symmetric Pascal matrices like that above, the sub- and superdiagonal matrices do not commute, so the (perhaps) tempting simplification involving the addition of the matrices cannot be made. A useful property of the sub- and superdiagonal matrices used for the construction is that both are nilpotent; that is, when raised to a sufficiently great integer power, they degenerate into the zero matrix. (See shift matrix for further details.) As the n × n generalised shift matrices we are using become zero when raised to power n, when calculating the matrix exponential we need only consider the first n + 1 terms of the infinite series to obtain an exact result. ==Variants== Interesting variants can be obtained by obvious modification of the matrix-logarithm PL 7 and then application of the matrix exponential. The first example below uses the squares of the values of the log-matrix and constructs a 7 × 7 "Laguerre"- matrix (or matrix of coefficients of Laguerre polynomials \begin{array}{lll} & LAG_7=\exp \left ( \left [ \begin{smallmatrix} . & . & . & . & . & . & . \\ 1 & . & . & . & . & . & . \\ . & 4 & . & . & . & . & . \\ . & . & 9 & . & . & . & . \\ . & . & . & 16 & . & . & . \\ . & . & . & . & 25 & . & . \\ . & . & . & . & . & 36 & . \end{smallmatrix} \right ] \right ) = \left [ \begin{smallmatrix} 1 & . & . & . & . & . & . \\ . & 1 & . & . & . & . & . \\ 1 & . & 1 & . & . & . & . \\ . & 3 & . & 1 & . & . & . \\ 3 & . & 6 & . & 1 & . & . \\ . & 15 & . & 10 & . & 1 & . \\ 15 & . & 45 & . & 15 & . & 1 \end{smallmatrix} \right ] \quad \end{array} If this matrix is inverted (using, for instance, the negative matrix-logarithm), then this matrix has alternating signs and gives the coefficients of the derivatives (and by extension the integrals) of Gauss' error-function. (Literature about generalizations to greater powers is not found yet.) Another variant can be obtained by extending the original matrix to negative values: \begin{array}{lll} & \exp \left ( \left [ \begin{smallmatrix} . & . & . & . & . & . & . & . & . & . & . & . \\ -5& . & . & . & . & . & . & . & . & . & . & . \\ . &-4 & . & . & . & . & . & . & . & . & . & . \\ . & . &-3 & . & . & . & . & . & . & . & . & . \\ . & . & . &-2 & . & . & . & . & . & . & . & . \\ . & . & . & . &-1 & . & . & . & . & . & . & . \\ . & . & . & . & . & 0 & . & . & . & . & . & . \\ . & . & . & . & . & . & 1 & . & . & . & . & . \\ . & . & . & . & . & . & . & 2 & . & . & . & . \\ . & . & . & . & . & . & . & . & 3 & . & . & . \\ . & . & . & . & . & . & . & . & . & 4 & . & . \\ . & . & . & . & . & . & . & . & . & . & 5 & . \end{smallmatrix} \right ] \right ) = \left [ \begin{smallmatrix} 1 & . & . & . & . & . & . & . & . & . & . & . \\ -5 & 1 & . & . & . & . & . & . & . & . & . & . \\ 10 & -4 & 1 & . & . & . & . & . & . & . & . & . \\ -10 & 6 & -3 & 1 & . & . & . & . & . & . & . & . \\ 5 & -4 & 3 & -2 & 1 & . & . & . & . & . & . & . \\ -1 & 1 & -1 & 1 & -1 & 1 & . & . & . & . & . & . \\ . & . & . & . & . & 0 & 1 & . & . & . & . & . \\ . & . & . & . & . & . & 1 & 1 & . & . & . & . \\ . & . & . & . & . & . & 1 & 2 & 1 & . & . & . \\ . & . & . & . & . & . & 1 & 3 & 3 & 1 & . & . \\ . & . & . & . & . & . & 1 & 4 & 6 & 4 & 1 & . \\ . & . & . & . & . & . & 1 & 5 & 10 & 10 & 5 & 1 \end{smallmatrix} \right ] . \end{array} ==See also== Pascal's triangle LU decomposition Riordan Array ==References== ==External links== G. Helms Pascalmatrix in a project of compilation of facts about Numbertheoretical matrices G. Helms Gauss-matrix Weisstein, Eric W. Gaussian-function Weisstein, Eric W. Erf-function Weisstein, Eric W. "Hermite Polynomial". Hermite-polynomials (Related to Gauss-matrix).

Text derived from https://en.wikipedia.org/w/index.php?oldid=1370501775 under CC BY-SA 4.0; modifications have been made.