Model Validation Basics

Ways to validate models, refine models, troubleshooting

After estimating a model, you can validate whether it reproduces the system behavior within acceptable bounds. You can then iterate between model refinement and validation until you find the simplest model that best captures the system dynamics.

Topics

Validating Models After Estimation
Introduction to validating models and supported model plots.
Refine Linear Parametric Models
Procedures for refining model parameters after estimating a model or constructing the model with initial parameter guesses.
Refine Initial ARMAX Model at Command Line
This example shows how to estimate an initial model and refine it using pem.
Refine ARMAX Model with Initial Parameter Guesses at Command Line
This example shows how to refine models for which you have initial parameter guesses.

Troubleshooting

Troubleshooting Model Estimation

Adjust your modeling strategy based on model-validation results.

Resolve Fit Value Differences Between Model Identification and compare Command

This example shows how NRMSE fit values computed by model identification functions and by the compare function can differ because of differences in initial conditions and prediction horizon settings.