User`s guide

8 Model Analysis
the estimated uncertainty in the model parameters to calculate confidence
intervals and assumes the estimates have a Gaussian distribution.
For example, for a 95% confidenceinterval,theregionaroundthenominal
curve represents the range of values that have a 95% probability of being the
true system response. You can specify the confidence interval as a probability
(between 0 and 1) or as the number of standard deviations of a Gaussian
distribution. For example, a probability of 0.99 (99%) corresponds to 2.58
standard deviations.
Note The calculation of the confidence interval assum es that the model
sufficiently describes the system dynamics and the m odel residuals pass
independence tests.
How to Plot and Compare Model Output at the
Command Line
You can plot simulated and predicted model output using the com par e, sim,
and
predict commands.
Simulation and prediction require input data, a model, a nd the values of
the initial states. If you estimated the model using one data set, but want
to simulate the model using a different data set, the initial states of your
simulation must be consistent with thedatayouuseforsimulation.
Note compare automatically estimates the initial states from the data and
ensures consistency.
By default, sim and predict use the initial states that were derived from the
datayouusedtoestimatethemodel. Theseinitialstatesarenotappropriate
if you are simulating or predicting output using new data.
To use
sim or predict with a data set that differs from the data yo u u sed to
estimate the model, first estimate the new initial states
X0est from the data
using
findstates:
8-14