Main Content

Preparing Time-Series Data

Before you can estimate models for time series data, you must import your data into the MATLAB® software. You can only use time domain data. For information about which variables you need to represent time series data, see Time-Series Data Representation.

For more information about preparing data for modeling, see Ways to Prepare Data for System Identification.

If your data is already in the MATLAB workspace, you can import it directly into the System Identification app. If you prefer to work at the command line, you must represent the data as a System Identification Toolbox™ data object instead.

In the System Identification app — When you import scalar or multiple-output time series data into the app, leave the Input field empty. For more information about importing data, see Represent Data.

At the command line — To represent a time series vector or a matrix s as an iddata object, use the following syntax:

y = iddata(s,[],Ts);

s contains as many columns as there are measured outputs and Ts is the sample time.