System Identification Toolbox™

arxdata

ARX parameters from multiple-output idarx or single-output idpoly objects with variance information

Syntax

[A,B] = arxdata(m)
[A,B,dA,dB] = arxdata(m)

Arguments

model

Name of the idarx model object, which belongs to the idmodel abstract class.

Note   Also accepts idpoly models with an underlying ARX structure with orders nc=nd=nf=0.

Description

[A,B] = arxdata(m) returns A and B as 3–D arrays.

Suppose ny is the number of outputs (the dimension of the vector y(t)) and nu is the number of inputs.

A is an ny-by-ny-by-(na+1) array such that

A(:,:,k+1) = Ak
A(:,:,1) = eye(ny)

where k=0,1,...,na.

B is an ny-by-nu-by-(nb+1) array with

B(:,:,k+1) = Bk

A(0) is always the identity matrix. The leading entries in B equal to zero, which means there are no delays in the model.

Note   For a time series, B = [].

[A,B,dA,dB] = arxdata(m) returns A and B matrices, and dA and dB as the estimated standard deviations of A and B, respectively.

Remarks

A and B are 2–D or 3–D arrays and are returned in the standard multivariable ARX format (see idarx), describing the model.

Here and

A and B matrices, which have dimensions ny-by-ny and ny-by-nu, respectively. Here, ny is the number of outputs (the dimension of the vector y(t)) and nu is the number of inputs.

See Also

arx

arxstruc