ParameterStatistics

Calculate parameter statistics for linear model

collapse all in page

    Syntax

    values = ParameterStatistics(linearmodel)
    values = ParameterStatistics(linearmodel,statType)

    Description

    values = ParameterStatistics(linearmodel) calculates the parameter statistics for linearmodel.

    values = ParameterStatistics(linearmodel,statType) calculates the parameter statistics using the specified statType.

    example

    Examples

    collapse all

    Calculate parameter statistics of knot model.

    values = ParameterStatistics(knot)
    values = 
          Alias: [7x3 double]
     Covariance: [7x7 double]
    Correlation: [7x7 double]
      VIFsingle: [5x5 double]
    VIFmultiple: [7x1 double]
     VIFpartial: [5x5 double]
       Stepwise: [10x4 double]

values.Stepwise
ans =
  1.0e+003 *
    0.0190    0.0079    0.0210       NaN
    0.0000    0.0000    0.0210    1.9801
    0.0000    0.0000    0.0200    0.2984
   -0.0000    0.0000    0.0200    0.2768
    0.0000    0.0000    0.0200    0.2890
   -0.0526    0.0367    0.0210    0.2679
    0.0911    0.0279    0.0210    0.3837
   -0.0041    0.0024    0.0210    0.2728
   -0.0178    0.0095    0.0200    0.2460
    0.0001    0.0000    0.0210    0.3246

    Input Arguments

    collapse all

    Model object, specified as a mbcmodel.linearmodel object.

    Type of parameter statistics, specified as a character vector specifying a particular statistic or a cell array of character vectors specifying a number of statistics to output. The valid data types are

    • 'Alias'

    • 'Covariance'

    • 'Correlation'

    • 'VIFsingle'

    • 'VIFmultiple'

    • 'VIFpartial'

    • 'Stepwise'

    All of these types (except 'Stepwise') appear in the Design Evaluation tool. See documentation for more details of these matrices.

    The Stepwise field contains the values found in the Stepwise table. In this array (and in the Stepwise GUI) you can see for each parameter in the model: the value of the coefficient, the standard error of the coefficient, the t value and Next PRESS ( the value of PRESS if the status of this term is changed at the next iteration). See the documentation for the Stepwise table. You can also see these Stepwise values when you use StepwiseRegression.

    Output Arguments

    collapse all

    Parameter statistics values, returned as a structure.

    If statType is a character vector, then values is an array of doubles. If statType is a cell array of character vectors, then values is a cell array of array of doubles.

    Version History

    Introduced before R2006a

    See Also