Learn more about Model-Based Calibration

ResponseFeatures(Local Model) - Set of response features for local model

Syntax

RFs = L.ResponseFeatures

Description

This is a property of the local model object, mbcmodel.localmodel.

RFs = L.ResponseFeatures returns a mbcmodel.responsefeatures object. L is the local model.

See Understanding Model Structure in the Getting Started documentation for an explanation of the relationships between local models, local responses, and other responses.

Available properties and methods are described in the following tables.

Property	Description
EvaluationPoints	Cell array of evaluation points for the response feature set (read-only). An element of `EvaluationPoints` is empty if the response feature does not use the Evaluation point. This property is set up when the response feature is created (see the Add method).
Types	Cell array of types for response feature set (read-only). This property is set up when the response feature is created (see the Add method).
NumberOfResponseFeatures	Number of response features in set (read-only).
IsFitted	The local model has been fitted.

Method	Description
Add	Add new response feature to response feature set RF = Add(RF,RFtype) `RFtype` is a description string belonging to the set of alternative response features. See getAlternativeTypes. RF = Add(RF,RFtype,EvaluationPoint) `EvaluationPoint` is a row vector with an element for each model input and is used for response features that require an input value to evaluate the response feature (e.g., function evaluation, derivatives). It is an error to specify an evaluation point for a response feature type that does not require an evaluation point.
Remove	Remove a response feature from the response feature set RF = Remove(RF,index)
Select	Select a subset of response features from the response feature set RF = Select(RF,indices)
getDefaultSet	List of default response features RF = getDefaultSet(RF) Returns an `mbcmodel.responsefeatures` object with the default set of response features for the local model.
getAlternativeTypes	List of all alternative response feature types for local model RFtypes = getAlternativeTypes(RF) Returns a cell array of response feature type strings for the local model.
Evaluate	Evaluate response features rfvals = Evaluate(RF); Returns the values for the response features for the current local model. [rfvals,stderr] = Evaluate(RF) Also returns the standard errors for the response features for the current local model. The local model must be fitted before evaluating response features.
Jacobian	Jacobian matrix of response features with respect to parameters J = Jacobian(RF) The local model must be fitted before calculating the Jacobian matrix.
Covariance	Covariance matrix for response features rfvals = Covariance(RF); The local model must be fitted before calculating the covariance matrix.
Correlation	Correlation matrix for response features rfvals = Correlation(RF) Errors occur if model is not fitted.
ReconstructSets	List of subsets of response features which can be used to reconstruct the local model RFlist = ReconstructSets(RF) `RFlist` is a cell array of `mbcmodel.responsefeatures`. Each element of `RFlist` can be used to reconstruct the local model from response feature values.

Examples

First, create a local model object:

L = mbcmodel.CreateModel('Local Polynomial',2)
 

L =
 
   1 + 2*X1 + 8*X2 + 3*X1^2 + 6*X1*X2 + 9*X2^2 + 4*X1^3...
 + 5*X1^2*X2 + 7*X1*X2^2 +
   10*X2^3                                                                       
   InputData: [0x2 double]
   OutputData: [0x1 double]
   Status: Not fitted
   Linked to Response: not linked

The properties of the local model object are the same as the properties of an mbcmodel.model object with the additional property "ResponseFeatures". Look at the response features property as follows:

>> RFs = L.ResponseFeatures
 
RFs =
 
Response features for Polynomial
    'Beta_1'
    'Beta_X1'
    'Beta_X1^2'
    'Beta_X1^3'
    'Beta_X1^2*X2'
    'Beta_X1*X2'
    'Beta_X1*X2^2'
    'Beta_X2'
    'Beta_X2^2'
    'Beta_X2^3'

% Set up response features
RFtypes = getAlternativeTypes(RFs);
RF = Add(RF, RFtypes{end},-10);

% assign to local model
L.ResponseFeatures = RFs;

Response

ResponseFeatures(Local Response)

Recommended Products

Get MATLAB trial software

Includes the most popular MATLAB recorded presentations with Q&A sessions led by MATLAB experts.

Get the Interactive Kit