ar_model

AR_MODEL compute AR-models parameters of input signal using Yule-Walker method.

LAMBDA = AR_MODEL(Y, N) estimates an N:th order autoregressive polynomial model (AR) for time series Y:

y(t) + l_1 * y(t-1) + l_2 * y(t-2) + ... +l_N * y(t-N) = e(t)

Inputs:

Y: The time series to be modeled, a column vector of values. The data must be uniformly sampled.

N: The order of the AR model (positive integer)

Output:

LAMBDA: AR model delivered as an array where are [1 l_1 l_2 l_3 ... l_N].
The model is estimated using "Yule-Walker" approach with no windowing.