Main Content

Nonlinear Regression

Least-squares estimation to fit grouped or pooled data, single or multiple experiments

Functions

fit Perform parameter estimation using SimBiology problem object
sbiofit Perform nonlinear least-squares regression
sbionlinfit Perform nonlinear least-squares regression usingSimBiologymodels (requiresStatistics and Machine Learning Toolboxsoftware)
sbioparamestim Perform parameter estimation
sbiosampleparameters Generate parameters by sampling covariate model (requiresStatistics and Machine Learning Toolboxsoftware)
sbiosampleerror Sample error based on error model and add noise to simulation data
sbioparameterci Compute confidence intervals for estimated parameters (requiresStatistics and Machine Learning Toolbox)
sbiopredictionci Compute confidence intervals for model predictions (requiresStatistics and Machine Learning Toolbox)

Objects

fitproblem SimBiology problem object for parameter estimation
groupedData Table-like collection of data and metadata
EstimatedInfo object Object containing information about estimated model quantities
LeastSquaresResults object 结果对象包含估计结果from least-squares regression
Observable Object containing expression for post-simulation calculations
OptimResults object Estimation results object, subclass ofLeastSquaresResults
NLINResults object Estimation results object, subclass ofLeastSquaresResults
ParameterConfidenceInterval Object containing confidence interval results for estimated parameters
PredictionConfidenceInterval Object containing confidence interval results for model predictions

Apps

SimBiology Model Builder Build QSP, PK/PD, and mechanistic systems biology models interactively
SimBiology Model Analyzer Analyze QSP, PK/PD, and mechanistic systems biology models

Examples and How To

App Workflow

Programmatic Workflow

Concepts

  • Nonlinear Regression

    The purpose of regression models is to describe a response variable as a function of independent variables.

  • Supported Methods for Parameter Estimation in SimBiology

    SimBiology®supports a variety of optimization methods for least-squares and mixed-effects estimation problems.

  • Error Models

    SimBiology supports the error models described in the following table.

  • Progress Plot

    The progress plot provides the live feedback on the status of parameter estimation while usingsbiofit,sbiofitmixed, or theFit Dataprogram in theSimBiology Model Analyzer应用程序。