Learn more about SimBiology

sbionlmefit - Estimate nonlinear mixed effects using SimBiology models

Syntax

results = sbionlmefit(modelObj, modelMapObject, pkDataObject, Beta0) results = sbionlmefit(modelObj, modelMapObject, pkDataObject, Beta0, ...) ... = sbionlmefit(..., optionstruc) [results, SimDataI, SimDataP] = sbionlmefit(...)

Arguments

Input Arguments

`modelObject`	SimBiology model object used to fit observed data.
`modelMapObject`	`PKModelMap` object that defines the roles of the model components used for estimation. For more information, see `PKModelMap object`.
`pkDataObject`	`PKData` object that defines the data to use in fitting and the roles of the columns used for estimation. `pkDataObject` must define target data for at least two groups. For more information, see `PKData object`.
`Beta0`	A vector of initial estimates for the fixed effects. The length of `Beta0` must equal `modelMapObject.Estimated` if you are not defining any additional covariates. The final estimates for the fixed and random effects for these parameters are log-transformed in the results. For more information on specifying initial estimates see Setting Initial Estimates in the SimBiology User's Guide.

Output Arguments

results

A structure with the following fields:

beta — Log-transformed estimates for the fixed effects
psi — Estimated covariance matrix of the random effects
stats — Structure with statistics such as AIC, BIC.
For a complete list, refer to nlmefit in the Statistics Toolbox User's Guide.
b — Log-transformed estimates of the random effects for each group in pkDataObject.

SimDataI SimDataI contains the data from simulating the model using the estimated parameter values, for individuals.

SimDataP SimDataP contains the data from simulating the model using the estimated parameter values, for the population.

Description

Note This function requires nlmefit in Statistics Toolbox (Version 7.0 or later).

results = sbionlmefit(modelObj, modelMapObject, pkDataObject, Beta0) performs nonlinear mixed effects estimation using the SimBiology model, modelObj and returns estimated results in the results structure.

results = sbionlmefit(modelObj, modelMapObject, pkDataObject, Beta0, ...) lets you supply additional arguments accepted by nlmefit. Use the FEGroupDesign property to specify the design matrix for each of the groups. See Specifying the Covariate Model in the SimBiology User's Guide documentation.

sbionlmefit does not support the following arguments:

FEConstDesign
FEObsDesign
FEParamsSelect
REConstDesign
REGroupDesign
REObsDesign
Vectorization
Jacobian

... = sbionlmefit(..., optionstruc) where the optionstruc is a struct that can contain fields and values where the field names are the parameter-names accepted by nlmefit and the field values are the associated parameter-values. SimBiology software contains a function (sbiofitstatusplot) that you can specify in the options structure. This function lets you monitor the status of fitting.

See nlmefit in the Statistics Toolbox User's Guide documentation for information on the accepted properties.

[results, SimDataI, SimDataP] = sbionlmefit(...) returns simulation data of the SimBiology model, modelObj, using the estimated values of the parameters.