LinearMixedModelOptions<T>

Options & Configuration · AiDotNet.Models.Options

Configuration options for mixed-effects (hierarchical/multilevel) models.

Inheritance: ModelOptions → RegressionOptions<T> → LinearMixedModelOptions<T>

For Beginners

Mixed-effects models are for data that has “groups” or “clusters”.

Examples of grouped data:

  • Students nested in schools
  • Patients nested in hospitals
  • Repeated measurements on individuals
  • Products sold in different stores

Why use mixed models instead of regular regression?

  1. Properly accounts for non-independence within groups
  2. Gives you valid standard errors and p-values
  3. Shares information across groups (partial pooling)
  4. Estimates how much variation is between vs. within groups

How It Works

Mixed-effects models handle data with natural grouping or clustering by estimating both population-level (fixed) effects and group-level (random) effects. They properly account for correlation within groups and provide valid inference.

Constructors

Constructor Summary
LinearMixedModelOptions()

Properties

Property Summary
MixedEffectsEstimationMethod EstimationMethod { get; set; } Gets or sets the estimation method.
MixedEffectsOptimizer Optimizer { get; set; } Gets or sets the optimizer to use.
bool BoundVarianceComponents { get; set; } Gets or sets whether to use the bounded optimization for variance components.
bool ComputeBLUPs { get; set; } Gets or sets whether to compute predicted random effects (BLUPs).
bool ComputeRSquared { get; set; } Gets or sets whether to compute marginal and conditional R-squared.
bool ComputeVarianceCI { get; set; } Gets or sets whether to compute confidence intervals for variance components.
bool UseSVD { get; set; } Gets or sets whether to use singular value decomposition for numerical stability.
bool Verbose { get; set; } Gets or sets whether to print verbose output during fitting.
double ConfidenceLevel { get; set; } Confidence level of the intervals reported for the fixed effects (for example 0.95).
double Tolerance { get; set; } Convergence tolerance of the variance-component estimation.
int MaxIterations { get; set; } Maximum number of iterations of the variance-component estimation.