Korovkin tests, approximation, and ergodic theory
Stefano Serra‐Capizzano · Mathematics of Computation · 2000
We consider sequences of s ⋅ k ( n ) × t ⋅ k ( n ) s\cdot k(n)\times t\cdot k(n) matrices { A n ( f ) } \{A_n(f)\} with a block structure spectrally distributed as an L 1 L_1 p p -variate s × t s\times t matrix-valued function f f , and, for any n n , we suppose that A n ( ⋅ ) A_n(\cdot ) is a linear and positive operator. For every fixed n n we approximate the matrix A n ( f ) A_n(f) in a suitable linear space M n \mathcal {M}_n of s ⋅ k ( n ) × t ⋅ k (