Time series: theory and methods
Time series: theory and methods
An efficient and versatile algorithm for computing the covariancefunction of an ARMA process
IEEE Transactions on Signal Processing
Hi-index | 0.00 |
A new formulation facilitating the determination of the autocorrelation lags, and consequently the autocorrelation matrix, of an autoregressive moving average signal of arbitrary order (ARMA(p, q)) is presented. The main features of our formulation, stated as a set of linear equations to be solved, are 1) its conceptual simplicity 2) its reliance only on primary parameters of the ARMA(p, q) model, and 3) the simplicity of the resulting linear equations.