# Estimate State-Space Equivalent of ARMAX and OE Models

This example shows how to estimate ARMAX and OE-form models using the state-space estimation approach.

You can estimate the equivalent of multiple-output ARMAX and Output-Error (OE) models using state-space model structures:

Convert the resulting models into `idpoly` models to see them in the commonly defined ARMAX or OE forms.

`load iddata1 z1`

Estimate state-space models.

```mss_noK = n4sid(z1,2,'DisturbanceModel','none'); mss = n4sid(z1,2);```

`mss_noK` is a second order state-space model with no disturbance model used during estimation. `mss` is also a second order state-space model, but with an estimated noise component. Both models use the measured data set `z1` for estimation.

Convert the state-space models to polynomial models.

```mOE = idpoly(mss_noK); mARMAX = idpoly(mss);```

Converting to polynomial models results in the parameter covariance information for `mOE` and `mARMAX` to be lost.

You can use one of the following to recompute the covariance:

• Zero-iteration update using the same estimation data.

• `translatecov` as a Gauss approximation formula-based translation of covariance of `mss_noK` and `mss` into covariance of `mOE` and `mARMAX`.

Reestimate `mOE` and `mARMAX` for the parameters of the polynomial model using a zero iteration update.

```opt = polyestOptions; opt.SearchOptions.MaxIterations = 0; mOE = polyest(z1,mOE,opt); mARMAX = polyest(z1,mARMAX,opt);```

The options object, `opt`, specifies a zero iteration update for `mOE` and `mARMAX`. Consequently, the model parameters remain unchanged and only their covariance information is updated.

Alternatively, you can use `translatecov` to convert the estimated models into polynomial form.

```fcn = @(x)idpoly(x); mOEt = translatecov(fcn,mss_noK); mARMAXt = translatecov(fcn,mss);```

Because `polyest` and `translatecov` use different computation algorithms, the covariance data obtained by running a zero-iteration update may not match that obtained using `translatecov`.

You can view the uncertainties of the model parameters using `present(mOE)` and `present(mARMAX)`.

You can use a state-space model with $K=0$ (Output-Error (OE) form) for initializing a Hammerstein-Wiener estimation at the command line. This initialization may improve the fit of the model. See Initialize Hammerstein-Wiener Estimation Using Linear Model.