Commands for Model Estimation

The following tables summarize System Identification Toolbox™ commands for offline and online estimation. For detailed information about using each command, see the corresponding reference page.

You can compile all the estimation commands using MATLAB^® Compiler™ software. Using MATLAB Coder™ software, you can only generate C and C++ code for online estimation commands, except for rpem, rplr, and segment.

Commands for Offline Estimation

Model Type	Estimation Commands
Transfer function models	tfest
Process models (low-order transfer functions expressed in time-constant form)	procest
Linear input-output polynomial models	armax (ARMAX and ARIMAX models) arx (ARX and ARIX models) bj (BJ only) iv4 (ARX only) ivx (ARX only) oe (OE only) polyest (for all models)
State-space models	n4sid ssest ssregest
Frequency-response models	etfe spa spafdr
Correlation models	cra impulseest
Linear time-series models	ar arx (for multiple outputs) ivar
Linear grey-box models	greyest
Nonlinear ARX models	nlarx
Hammerstein-Wiener models	nlhw
Nonlinear grey-box models	nlgreyest
Linear and nonlinear models	pem

Commands for Online Estimation

Model Type	Estimation Commands
Linear input-output polynomial models	recursiveARX recursiveARMAX recursiveOE recursiveBJ
Linear time-series models	recursiveAR recursiveARMA
Model that is linear in parameters	recursiveLS
Linear polynomial models	rpem rplr segment (AR, ARMA, ARX, and ARMAX models only)