Struct BayesianLinearRegression 

pub struct BayesianLinearRegression {
    pub prior_mean: Array1<f64>,
    pub prior_precision: Array2<f64>,
    pub prior_alpha: f64,
    pub prior_beta: f64,
    pub fit_intercept: bool,
}

Bayesian linear regression with normal-inverse-gamma prior

This implements Bayesian linear regression where:

• Coefficients have a normal prior
• Noise variance has an inverse-gamma prior
• Posterior is analytically tractable

Fields

prior_mean: Array1<f64>

Prior mean for coefficients

prior_precision: Array2<f64>

Prior precision matrix for coefficients

prior_alpha: f64

Prior shape parameter for noise variance

prior_beta: f64

Prior scale parameter for noise variance

fit_intercept: bool

Whether to include intercept

Implementations

impl BayesianLinearRegression

pub fn new(n_features: usize, fit_intercept: bool) -> StatsResult<Self>

Create a new Bayesian linear regression model

pub fn with_priors( prior_mean: Array1<f64>, prior_precision: Array2<f64>, prior_alpha: f64, prior_beta: f64, fit_intercept: bool, ) -> StatsResult<Self>

Create with custom priors

pub fn fit( &self, x: ArrayView2<'_, f64>, y: ArrayView1<'_, f64>, ) -> StatsResult<BayesianRegressionResult>

Fit the Bayesian regression model

pub fn predict( &self, x: ArrayView2<'_, f64>, result: &BayesianRegressionResult, ) -> StatsResult<BayesianPredictionResult>

Make predictions on new data

pub fn predict_with_credible_interval( &self, x: ArrayView2<'_, f64>, result: &BayesianRegressionResult, confidence: f64, ) -> StatsResult<BayesianPredictionResult>

Compute credible intervals for predictions

Trait Implementations

impl Clone for BayesianLinearRegression

fn clone(&self) -> BayesianLinearRegression

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for BayesianLinearRegression

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.