Matrices Associated to Biindexed Linear Recurrence Relations

Abstract

The factorization of matrix \(A\) with entries \(a_{i,j}\) determined by \(a_{i,j} = \alpha a_{i-1,j-1} + \beta a_{i,j-1}\) is derived as \(A = TP^T\). An interesting factorization of matrix \(B\) with entries \(b_{i,j} = \alpha b_{i-1,j} + \beta b_{i,j-1}\) is given by \(B = P[\alpha]TP^{T}[\beta]\). The beautiful factorization of matrix \(C\) whose entries satisfy \(c_{i,j} = \alpha c_{i-1,j} + \beta c_{i-1,j-1} + Ye_{i-1,j-1}\) is founded to be \(C = P[\alpha]DP^T[\beta]\), where \(T\) is a Toeplitz matrix, and \(P\) and \(P[\alpha]\) are Pascal matrices. The matrix product factorization to the problem is solved perfectly so far.