Main Content

getpvec

Obtain model parameters and associated uncertainty data

    Description

    example

    params = getpvec(sys) returns the parameters values of the identified model sys.

    example

    [params,stdDev] = getpvec(sys) also returns the standard deviation value for the parameters.

    example

    [___] = getpvec(sys,'free') returns values for only the free parameters of sys. For idnlarx and idnlhw models, all parameters are treated as free.

    Examples

    collapse all

    Load the estimation data.

    load iddata1 z1;

    Estimate a transfer function model.

    sys = tfest(z1,3);

    Retrieve the parameter values from the estimated model.

    pvec = getpvec(sys);

    Load the estimation data.

    load iddata2 z2;

    Estimate a state-space model.

    sys = ssest(z2,3);

    Retrieve the model parameters, pvec, and associated standard deviations, pvec_sd, from the estimated model.

    [pvec,pvec_sd] = getpvec(sys);

    Load the estimation data.

    load iddata2 z2;

    Estimate a state-space model.

    sys = ssest(z2,3);

    Retrieve the values of the free parameters from the estimated model.

    pvec = getpvec(sys,'free');

    Input Arguments

    collapse all

    Identified model, specified as one of the following model objects or an array of such objects.

    Output Arguments

    collapse all

    Parameter values, returned as a vector.

    If sys is a model array, then params is a cell array of parameter value vectors corresponding to each model in sys.

    params is [] for idnlarx and idnlhw models that have not been estimated.

    Standard deviation values for parameters, returned as a vector.

    If sys is a model array, then stdDev is a cell array of standard deviation value vectors corresponding to each model in sys.

    If the model covariance information for sys is not available, then stdDev is []. As a result, stdDev is always [] for idnlarx and idnlhw models because these models do not store parameter covariance information.

    Version History

    Introduced in R2012a