Skip to main content

loo_cv_press

Function loo_cv_press 

Source
pub fn loo_cv_press(
    fit: &FregreLmResult,
    data: &FdMatrix,
    y: &[f64],
    scalar_covariates: Option<&FdMatrix>,
) -> Result<LooCvResult, FdarError>
Expand description

LOO-CV / PRESS diagnostics for a linear functional regression model.

Uses the hat-matrix shortcut: LOO residual = e_i / (1 - h_ii).

ยงErrors

Returns FdarError::InvalidDimension if data has zero rows, its column count does not match fit.fpca.mean, y.len() does not match the row count, or the number of rows is not greater than the number of model parameters. Returns FdarError::ComputationFailed if Cholesky factorization fails or the total sum of squares is zero.