Estimate parameters of ARMAX model using timedomain data
sys = armax(data,[na
nb nc nk])
sys = armax(data,[na
nb nc nk],Name,Value)
sys = armax(data,init_sys)
sys = armax(data,___,opt)
Note:

returns an sys
= armax(data
,[na
nb nc nk]
)idpoly
model, sys
,
with estimated parameters and covariance (parameter uncertainties).
Estimates the parameters using the predictionerror method and specified
polynomial orders.
returns
an sys
= armax(data
,[na
nb nc nk]
,Name,Value
)idpoly
model, sys
, with
additional options specified by one or more Name,Value
pair
arguments.
estimates
a polynomial model using the ARMAX structure polynomial model sys
= armax(data
,init_sys
)init_sys
to
configure the initial parameterization.
specifies
estimation options using the option set sys
= armax(data
,___,opt
)opt
.

Estimation data. Specify You cannot use frequencydomain data for estimating ARMAX models. 

Polynomial orders.


Linear polynomial model that configures the initial parameterization
of
Use the To specify an initial guess for, say, the A(q)
term of To specify constraints for, say, the B(q)
term of
You can similarly specify the initial guess and constraints for the other polynomials. If 

Estimation options.
Use 
Specify optional commaseparated pairs of Name,Value
arguments.
Name
is the argument
name and Value
is the corresponding
value. Name
must appear
inside single quotes (' '
).
You can specify several name and value pair
arguments in any order as Name1,Value1,...,NameN,ValueN
.

Input delays. For a system with Default: 0 for all input channels 

Transport delays. Specify transport delays as integers denoting delay of a multiple
of the sample time For a MIMO system with Default: 0 for all input/output pairs 

Logical vector specifying integrators in the noise channel.
Setting $$A(q)y(t)=B(q)u(tnk)+\frac{C(q)}{1{q}^{1}}e(t)$$ Where, $$\frac{1}{1{q}^{1}}$$ is the integrator in the noise channel, e(t). Use For example, load iddata9 z9; z9.y = cumsum(z9.y); %integrated data sys = armax(z9,[4 1],'IntegrateNoise',true); compare(z9,sys,10) %10step ahead prediction Default: 

Identified ARMAX structure polynomial model.

armax
does not support continuoustime
model estimation. Use tfest
to
estimate a continuoustime transfer function model, or ssest
to estimate a continuoustime statespace
model.
Ljung, L. System Identification: Theory for the User, Upper Saddle River, NJ, PrenticeHal PTR, 1999. See chapter about computing the estimate.