When a linear model provides an insufficient description of the dynamics, you can try estimating a nonlinear models. To learn more about when to estimate nonlinear models, see Building Models from Data in the Getting Started Guide.
Before you can estimate models for time-series data, you must have already prepared the data as described in Preparing Time-Series Data.
For black-box modeling of time-series data, the toolbox supports nonlinear ARX models. To learn how to estimate this type of model, see Identifying Nonlinear ARX Models.
If you understand the underlying physics of the system, you can specify an ordinary differential or difference equation and estimate the coefficients. To learn how to estimate this type of model, see Estimating Nonlinear Grey-Box Models. See also Estimating Coefficients of ODEs to Fit Given Solution for an example of time series modeling using the grey-box approach.
For more information about validating models, see Validating Models After Estimation.