BetaRegressionOptions

Options & Configuration · AiDotNet.Models.Options

Configuration options for Beta Regression models.

For Beginners

Regular regression can predict any number, but what if you’re predicting something that must be between 0 and 1, like:

  • Percentage of students passing an exam
  • Proportion of defective products
  • Probability estimates
  • Market share percentages

Beta Regression handles this naturally by modeling the data as following a Beta distribution, which is naturally bounded between 0 and 1.

Key benefits:

  • Predictions are always in valid range (0,1)
  • Can model varying precision (some predictions more certain than others)
  • Handles skewed proportions well

How It Works

Beta Regression is used when your target variable is a proportion or rate bounded between 0 and 1 (exclusive). Examples include percentages, rates, and proportions.

Constructors

Constructor Summary
BetaRegressionOptions()

Properties

Property Summary
BetaLinkFunction LinkFunction { get; set; } Gets or sets the link function for the mean model.
bool CompressBoundaryValues { get; set; } When true, target values of exactly 0 or 1 are compressed into the open interval (0,1) using the Smithson and Verkuilen (2006) transformation rather than being rejected.
bool ModelVariablePrecision { get; set; } Gets or sets whether to model the precision parameter as a function of features.
bool UseRegularization { get; set; } Gets or sets whether to use regularization.
double LearningRate { get; set; } Gets or sets the learning rate for optimization.
double RegularizationStrength { get; set; } Gets or sets the regularization strength.
double Tolerance { get; set; } Gets or sets the convergence tolerance.
int MaxIterations { get; set; } Gets or sets the maximum number of iterations for optimization.
int? Seed { get; set; } Gets or sets the random seed for reproducibility.