Struct Blp 

pub struct Blp { /* private fields */ }

A fitted best linear predictor, one OLS fit per repetition of the signal.

Implementations

impl Blp

pub fn fit( signal: ArrayView2<'_, f64>, basis: ArrayView2<'_, f64>, ) -> Result<Self>

Regress each column of signal (shape (n, n_rep)) on basis (shape (n, p)).

pub fn fit_gate( signal: ArrayView2<'_, f64>, groups: ArrayView2<'_, f64>, ) -> Result<Self>

GATEs: BLP on mutually exclusive 0/1 group dummies.

pub fn is_gate(&self) -> bool

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

Basis coefficients (for GATEs: the group effects).

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

HC0 standard errors of the coefficients.

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

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

pub fn covariance(&self, rep: usize) -> Option<&Array2<f64>>

HC0 covariance of the coefficients for one repetition.

pub fn confint(&self, level: f64) -> Result<Array2<f64>>

Pointwise intervals for the coefficients, shape (p, 2).

pub fn predict( &self, basis: ArrayView2<'_, f64>, ) -> Result<(Array1<f64>, Array1<f64>)>

Predicted effects b(x)ᵀβ̂ and standard errors at new basis rows.

pub fn confint_at( &self, basis: ArrayView2<'_, f64>, level: f64, ) -> Result<Array2<f64>>

Pointwise intervals for τ(x) at new basis rows: columns [lower, effect, upper].

pub fn joint_confint_at( &self, basis: ArrayView2<'_, f64>, level: f64, n_boot: usize, seed: u64, ) -> Result<Array2<f64>>

Simultaneous intervals over all rows of basis, with critical values from a Gaussian multiplier bootstrap of b(x)ᵀ Ω^{1/2} Z / se(x).

Trait Implementations

impl Clone for Blp

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Blp

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

Formats the value using the given formatter.