User`s guide

Using Residual Analysis Plots to Validate M odels
describe how the output is formed from the corresponding input. For example,
when there is a peak outside the confidence interval for lag k,thismeansthat
the contribution to the output y(t) that or iginates from the input u(t-k) is
not properly described by the model. The models
arxqs and amx 2222 extend
beyond the confidence interval and do not perform as well as the other models.
Validating Models Using Analyzing R esiduals
To remove models with poor performance from the Residual Analysis plot,
click the model icons
arxqs, n4s3, arx223,andamx2222 in the System
Identification Tool GUI.
The Residual Analysis p lot now includes on ly the three models that pass the
residual tests:
arx692, n4s6,andamx3322.
The plo
ts for these models fall within the confidence intervals. Thus, when
choosi
ng the best model among several estimate d models, it is reasonable to
pick
a
mx3322
because it is a sim pler, low-order model.
8-23