Struct Rlm 

pub struct Rlm<N: RobustNorm> { /* private fields */ }

A robust linear model awaiting estimation.

Construct with Rlm::new, optionally configure the fit, and call Rlm::fit.

Implementations

impl<N: RobustNorm> Rlm<N>

pub fn new(endog: Array1<f64>, exog: Array2<f64>, norm: N) -> Result<Self>

A robust model with endog = y, exog = X, and criterion norm.

Defaults match the reference: scale_est='mad', conv='dev', update_scale=true, maxiter=50, tol=1e-8.

pub fn scale_est(self, s: ScaleEst) -> Self

Select the scale estimator (default ScaleEst::Mad).

pub fn conv(self, c: Conv) -> Self

Select the convergence criterion (default Conv::Deviance).

pub fn update_scale(self, u: bool) -> Self

Whether to re-estimate the scale every iteration (default true).

pub fn maxiter(self, m: usize) -> Self

Maximum number of IRLS iterations (default 50).

pub fn tol(self, t: f64) -> Self

Convergence tolerance on the criterion (default 1e-8).

pub fn nobs(&self) -> usize

Number of observations.

pub fn fit(&self) -> Result<RlmResults>

Estimate the model by IRLS.

Trait Implementations

impl<N: Clone + RobustNorm> Clone for Rlm<N>

fn clone(&self) -> Rlm<N>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<N: Debug + RobustNorm> Debug for Rlm<N>

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

Formats the value using the given formatter.