Struct AftFit 

pub struct AftFit { /* private fields */ }

A fitted accelerated failure time (AFT) parametric survival model,

log Tᵢ = xᵢᵀβ + σ·Wᵢ,

where W has the standard AftDistribution. Unlike the semiparametric Cox model this specifies the full baseline, so it yields absolute time predictions (e.g. median survival) and coefficients read as log time-acceleration: exp(βⱼ) > 1 multiplies survival time. Fit by maximum likelihood over right-censored data (Nelder–Mead on the log-likelihood, with standard errors from the numerical observed information).

Implementations

impl AftFit

pub fn new( time: Array1<f64>, event: Array1<f64>, x: Array2<f64>, dist: AftDistribution, ) -> Result<Self>

Fit an AFT model of survival (time, event) on X under dist.

event[i] is 1.0 for an observed event, 0.0 for right-censoring. X carries the fixed effects including an intercept (the model has a location term). Returns the standard-error covariance for the β coefficients (the scale is reported separately).

Errors

• RegressionError::EmptyInput / RegressionError::ShapeMismatch.
• RegressionError::InvalidResponse for non-positive times or event flags outside {0, 1}, or no events.
• RegressionError::NotConverged if the optimizer stalls.

pub fn distribution(&self) -> AftDistribution

The assumed distribution.

pub fn n_observations(&self) -> usize

Number of observations.

pub fn n_parameters(&self) -> usize

Number of coefficients (design columns).

pub fn coefficients(&self) -> ArrayView1<'_, f64>

AFT coefficients β (log time-acceleration scale), aligned to the design columns.

pub fn scale(&self) -> f64

The scale parameter σ (1 for the exponential).

pub fn covariance(&self) -> ArrayView2<'_, f64>

Coefficient covariance (inverse observed information).

pub fn log_likelihood(&self) -> f64

Maximized log-likelihood.

pub fn aic(&self) -> f64

AIC, −2ℓ + 2k, counting the scale parameter unless it is fixed.

pub fn coefficient_standard_errors(&self) -> Array1<f64>

Coefficient standard errors √diag(cov).

pub fn z_values(&self) -> Array1<f64>

Wald z-statistics βⱼ / seⱼ.

pub fn p_values(&self) -> Array1<f64>

Two-sided Wald p-values from the standard normal.

pub fn predict_median(&self, x: ArrayView2<'_, f64>) -> Array1<f64>

Predicted median survival time for a new design matrix x: exp(xβ + σ·median(W)).

Trait Implementations

impl Clone for AftFit

fn clone(&self) -> AftFit

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AftFit

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

Formats the value using the given formatter.