Function ridge_prediction_intervals 

pub fn ridge_prediction_intervals(
    fit: &RidgeFit,
    x_vars: &[Vec<f64>],
    new_x: &[&[f64]],
    alpha: f64,
) -> Result<PredictionIntervalOutput>

Computes approximate prediction intervals for Ridge regression.

Uses the conservative approximation with leverage from unpenalized X’X, MSE from the ridge fit, and effective degrees of freedom from fit.df.

Arguments

• fit - Reference to the fitted Ridge model
• x_vars - Original training predictor variables (each inner vec is one variable)
• new_x - New predictor values (each inner slice is one variable)
• alpha - Significance level (e.g., 0.05 for 95% prediction interval)