User`s guide

Using Residual Analysis Plots to Validate M odels
How to Plot Resid
uals at the Command Line
The following ta
ble summarizes commands that generate residual-analysis
plots for linear
and nonlinear models. For detailed information about this
command, see the
corresponding reference page.
Note Apply pe and
resid to one model at a time.
Command
Description
Exampl e
pe
Computes and plots model
prediction errors.
To plot the p red
iction
errors for the m
odel
model
using data data
,typethe
followin g com
mand:
pe(model,data)
resid
Performs whiteness a nd
independence tests on model
residuals, or prediction
errors. Uses validation data
input as m odel input.
To plot residu
al correlations
for the model
m
odel
using
data
data,ty
pe the following
command:
resid(model,data)
Example – Ex
amining Model Residuals
This example
shows how you can use residual analysis to evaluate model
quality.
Creating Residual Plots
1 To load the sampl e System Identi fi catio n Tool session that contains
estimated models, type the following command in the MATLAB C ommand
Window:
ident('dryer2_linear_models')
2 To generate a residual analysis plot, select the Model resids check box in
the System Identification Tool GUI.
8-21