ADIM adaptive information matrix. Estimates the inverse correlation matrix in an adaptive way. [IR, CC] = adim(U, UC [, IR0 [, CC0]]); U input data UC update coefficient 0 < UC << 1 IR0 initial information matrix CC0 initial correlation matrix IR information matrix (inverse correlation matrix) CC correlation matrix The algorithm uses the Matrix Inversion Lemma, also known as "Woodbury's identity", to obtain a recursive algorithm. IR*CC - UC*I should be approx. zero. Reference(s): [1] S. Haykin. Adaptive Filter Theory, Prentice Hall, Upper Saddle River, NJ, USA pp. 565-567, Equ. (13.16), 1996.
Package: tsa