The representations of the Drazin inverse of differences of two matrices (Q988620): Difference between revisions

In recent years, the representations and perturbation analysis of the Drazin inverse for matrices or operators have been investigated. These investigations motivated the authors of the present paper to study how to express the Drazin inverse of sums, differences and products of two matrices \(P\) and \(Q\). Let \(A\in\mathbb{C}^{n\times n}\). The Drazin inverse, denoted by \(A^D\), of matrix \(A\) is defined as the unique matrix satisfying \(A^{k+1}A^D= A^k\), \(A^DAA^D= A^D\), \(AA^D= A^D A\), where \(k=\text{Ind}(A)\) is the index of \(A\). In particular, if \(\text{Ind}(A)\leq 1\), then \(A^D\) is called the group inverse of \(A\) and is denoted by \(A^{\#}\). The paper gives explicit representations of \((P\pm Q)^D\), \((P\pm PQ)^D\) and \((PQ)^{\#}\) of two matrices \(P\) and \(Q\), as a function of \(P\), \(Q\), \(P^D\) and \(Q^D\), under the conditions \(P^3Q= QP\) and \(Q^3P= PQ\). The authors also clearly present the upper end lower bound of \(\| P^D- (P-Q)^D\|_2\).