System Identification    

Command Reference


This section contains detailed descriptions of all of the functions of user interest in the System Identification Toolbox. It begins with a list of functions grouped by subject area and continues with the entries in alphabetical order .

Information is also available through the on-line Help facility. By typing a function name without arguments, you also get immediate syntax help about its arguments for most functions

For ease of use, most functions have several default arguments. The Syntax first lists the function with the necessary input arguments and then with all the possible input arguments. The functions can be used with any number of arguments between these extremes. The rule is that missing, trailing arguments are given default values, as defined in the manual. Default values are also obtained by entering the arguments as the empty matrix [ ].

MATLAB does not require that you specify all of the output arguments; those not specified are not returned. For functions with several output arguments in the System Identification Toolbox, missing arguments are, as a rule, not computed, in order to save time.

Functions listed by Category

Help Functions
help ident

Lists the commands.

idhelp

A micro-manual.

idprops,
help idprops

Lists and explains the object properties.

The Graphical User Interface
ident

Open the interface.

midprefs

Set directory where to store start-up information.

Simulation and Prediction
idinput

Generate input signals.

pe

Compute prediction errors.

predict

Compute predictions according to model.

sim

Simulate a general linear system.

Data Manipulation
detrend

Remove trends from data.

get/set

Retrieve and modify iddata properties.

iddata

Construct a data object.

idfilt

Filter data.

merge (iddata)

Merge data sets into a multiple experiment set.

misdata

Reconstruct missing input and output data.

plot (iddata)

Plot data.

resample

Resample data.

Nonparametric Estimation
covf

Estimate covariance function.

cra

Estimate impulse response and covariance functions using correlation analysis.

impulse, step

Estimate impulse and step responses using high order parametric models.

etfe

Estimate spectra and transfer functions using direct Fourier techniques.

spa

Estimate spectra and transfer functions using spectral analysis.

Parameter Estimation
ar

Estimate AR model.

armax

Estimate ARMAX model.

arx

Estimate ARX model using least squares.

bj

Estimate Box-Jenkins model.

ivar

Estimate AR model using instrumental variable methods.

iv4

Estimate ARX model using four-stage instrumental variable method.

oe

Estimate Output-Error model.

n4sid

Estimate state-space model using subspace method.

pem

Estimate general linear model.

Model Structure Creation
idarx

Create multivariable ARX-models.

idfrd

Create Identified Frequency Response Data object.

idgrey

Create a greybox linear model using an M-file that you write.

idpoly

Create a model structure for input-output models defined as numerator and denominator polynomials.

idss

Create model structure for linear state-space models with known and unknown parameters.

Manipulating Model Structures  
get/set

Retrieve and modify model structures.

init

Select or randomize initial parameter values.

merge (idmodel)

Merge estimated models.

Model Conversions
arxdata

Compute ARX parameters.

idmodred

Reduce a model to lower order.

c2d

Transform from continuous to discrete time.

d2c

Transform from discrete to continuous time.

freqresp

Compute frequency response.

idfrd

Convert idmodel to the IDFRD object that contains frequency functions and spectra.

noisecnv

Convert noise inputs to measured channels

polydata

Compute transfer function polynomials.

ssdata

Compute state-space matrices.

tfdata

Compute transfer functions.

ss, tf, zpk, frd

Conversion of idmodel to the LTI-objects of the Control Systems Toolbox.

zpkdata

Compute zeros, poles, and gains.

Model Analysis  
bode

Plot Bode diagrams.

compare

Compare measured and simulated outputs.

ffplot

Plot frequency functions and spectra.

impulse, step

Plot impulse and step responses.

nyquist

Plot Nyquist diagrams.

present

Display model on screen.

pzmap

Plot zeros and poles.

view

Plot model characteristics using the LTI Viewer in the Control System Toolbox.

Model Validation
aic, fpe

Compute model selection criteria

arxstruc, selstruc

Select ARX-structure

compare

Compare model's simulated or predicted output with actual output

pe

Compute prediction errors

predict

Predict future outputs

resid

Compute and test model residuals

sim

Simulate a model

Assessing Model Uncertainty
simsd

Simulate responses from several possible models

bode, nyquist

Frequency responses with confidence regions

impulse, step, sim

Time responses with confidence regions

pzmap

Pole/zero plot with confidence regions

arxdata, polydata, ssdata, tfdata, zpkdata

Model data with variance information

Model Structure Selection
arxstruc

Compute loss functions for sets of ARX model structure.

ivstruc

Compute loss functions for sets of output error model structures.

n4sid, pem

State-space model order can be give as a range.

selstruc

Select structure.

struc

Generate sets of structures.

Recursive Parameter Estimation
rarmax

Estimate ARMAX or ARMA models recursively.

rarx

Estimate ARX or AR models recursively.

rbj

Estimate Box-Jenkins models recursively.

roe

Estimate Output-Error models (IIR-filters) recursively.

rpem

Estimate general input-output models using a recursive prediction error method.

rplr

Estimate general input-output models using a recursive pseudo-linear regression method.

segment

Segment data and estimate models for each segment.

General
get

Retrieve object properties.

set

Set object properties.

setpname

Set default, mnemonic parameter names.

size

Give sizes of the different objects.

timestamp

Show object's time of creation.


 Reference Functions Listed Alphabetically