System Identification Toolbox    

ARX Models

To estimate the parameters and of the ARX model (3-14), use the function arx.

Here na, nb, and nk are the corresponding orders and delays in (3-15) on page 3-11 that define the exact model structure. The function arx implements the least squares estimation method, using QR-factorization for overdetermined linear equations.

An alternative is to use the Instrumental Variable (IV) method described in connection with (3-39). This is obtained with

which gives an automatic (and approximately optimal) choice of the filters N and M in (3-39). (See the procedure (15.21)-(15.26) in Ljung (1999).)

Both arx and iv4 are applicable to arbitrary multivariable systems. If you have ny outputs and nu inputs, the orders are defined accordingly: na is an ny-by-ny matrix whose i -j entry gives the order of the polynomial that relates past values of to the current value of (i.e., past values of up to are used when predicting . Similarly, the i -j entries of the ny-by-nu matrices nu and nk, respectively, give the order and delay from input number j when predicting output number i. (See Multivariable ARX Models: The idarx Model and Function Reference for exact details.)


  Parametric Model Estimation AR Models