Alternative Model Exploration

Build and illustrate alternative models; select best models

When you use the MBC Model Fitting app to fit statistical models to experimental data, the app automatically fits models to your data. After assessing the initial model fit, use the app tools to create more models to search for the best fit. For more information, see Guidelines for Selecting the Best Model Fit.

Apps

MBC Model Fitting

Create experimental designs and statistical models for model-based calibration

Functions

expand all

Hierarchical Response Methods

`xregstatsmodel`	Class for evaluating models and calculating prediction error variance (PEV)
`AlternativeModelStatistics`	Summary statistics for alternative models
`CreateAlternativeModels`	Create alternative models from model template
`DoubleInputData`	Data being used as input to model
`DoubleResponseData`	Data being used as output to model for fitting
`Export`	Make command-line or Simulink export model
`OutlierIndices`	Indices of input data for response marked as outliers
`pev`	Predicted error variance of model at specified inputs
`PredictedValue`	Predicted value of model at specified inputs
`Remove`	Remove project model
`SummaryStatistics`	Summary statistics for response
`ValidationRMSE`	Calculates the validation RMSE for model data

Local Models

Local Response Methods

`mbcPointByPointModel`	Class for evaluating point-by-point models and calculating prediction error variance (PEV)
`AlternativeModelStatistics`	Summary statistics for alternative models
`CreateAlternativeModels`	Create alternative models from model template
`CreateResponseFeature`	Create new response feature for local model
`DiagnosticStatistics`	Diagnostic statistics for responses and local responses
`DoubleInputData`	Data being used as input to model
`DoubleResponseData`	Data being used as output to model for fitting
`Export`	Make command-line or Simulink export model
`MakeHierarchicalResponse`	Build two-stage model from response feature models
`ModelForTest`	Model for specified test
`OutlierIndices`	Indices of input data for response marked as outliers
`OutlierIndicesForTest`	Indices marked as outliers for test

Local Response Methods

`xregstatsmodel`	Class for evaluating models and calculating prediction error variance (PEV)
`pev`	Predicted error variance of model at specified inputs
`PEVForTest`	Local model predicted error variance for test
`PredictedValue`	Predicted value of model at specified inputs
`PredictedValueForTest`	Predicted values for test
`Remove`	Remove test plan, model, or boundary model
`RemoveOutliers`	Remove outliers in input data by index or rule, and refit models
`RemoveOutliersForTest`	Remove outliers on test by index or rule and refit models
`RestoreData`	Restore removed outliers
`RestoreDataForTest`	Restore removed outliers for test
`SummaryStatistics`	Summary statistics for response
`SummaryStatisticsForTest`	Statistics for specified test
`ValidationRMSE`	Calculates the validation RMSE for model data
`UpdateResponseFeatures`	Refit response feature models

Response Models

`xregstatsmodel`	Class for evaluating models and calculating prediction error variance (PEV)
`EvalModel`	Evaluate model
`ceval`	Evaluate boundary model
`dferror`	Degrees of freedom for model
`predint`	Confidence interval for model prediction
`nfactors`	Confidence interval for model prediction
`AlternativeModelStatistics`	Summary statistics for alternative models
`ChooseAsBest`	Choose best model from alternative responses
`CreateAlternativeModels`	Create alternative models from model template
`DiagnosticStatistics`	Diagnostic statistics for responses and local responses
`DoubleInputData`	Data being used as input to model
`DoubleResponseData`	Data being used as output to model for fitting
`Export`	Make command-line or Simulink export model
`OutlierIndices`	Indices of input data for response marked as outliers
`pev`	Predicted error variance of model at specified inputs
`PredictedValue`	Predicted value of model at specified inputs
`Remove`	Remove project model
`RemoveOutliers`	Remove outliers in input data by index or rule, and refit models
`RestoreData`	Restore removed outliers
`SummaryStatistics`	Summary statistics for response
`ValidationRMSE`	Calculates the validation RMSE for model data

Boundary Models

Boundary Classes

`mbcboundary.Boolean`	Boolean boundary model class
`mbcboundary.Model`	Boundary model class
`mbcboundary.PointByPoint`	Point-by-point boundary model class
`mbcboundary.Tree`	Root boundary tree class
`mbcboundary.TwoStageTree`	Root boundary tree class in two-stage test plans
`mbcboundary.TwoStage`	Two-stage boundary model class

Model Methods

`CreateBoundary`	Create boundary model
`designconstraint`	Convert boundary model to design constraint
`Evaluate`	Evaluate model, boundary model, or design constraint
`fit`	Fit model or boundary model to new or existing data, and provide summary statistics
`getAlternativeTypes`	Alternative boundary models

Boolean Methods

`CreateBoundary`	Create boundary model
`designconstraint`	Convert boundary model to design constraint
`Evaluate`	Evaluate model, boundary model, or design constraint
`getAlternativeTypes`	Alternative boundary models

PointByPoint Methods

`CreateBoundary`	Create boundary model
`designconstraint`	Convert boundary model to design constraint
`Evaluate`	Evaluate model, boundary model, or design constraint
`getAlternativeTypes`	Alternative boundary models

TwoStage Methods

`CreateBoundary`	Create boundary model
`designconstraint`	Convert boundary model to design constraint
`Evaluate`	Evaluate model, boundary model, or design constraint
`getAlternativeTypes`	Alternative boundary models
`getLocalBoundary`	Get local boundary model at operating point

Tree Methods

`Add`	Add boundary model to tree and fit to test plan data
`CreateBoundary`	Create boundary model
`Remove`	Remove test plan, model, or boundary model
`Update`	Update boundary model in tree and fit to test plan data

Topics

Create Alternative Models to Compare
Use the Model Template dialog box to add multiple model types to compare.
Compare Alternative Models
Use model plots and statistics to compare alternative models.
Explore Local Model Types
Set up local model types, including descriptions and illustrations of model shapes.
Explore Global Model Types
Set up global model types, including descriptions and illustrations of model shapes.
Explore Boundary Model Types
Construct and use boundary models.
Add Response Models and Datum Models
Add new response models, and use datum models.
Radial Basis Functions for Model Building
Build, tune, and evaluate radial basis function (RBF) models.