Struct Apos 

pub struct Apos { /* private fields */ }

Average potential outcomes of several treatment levels, fitted on shared sample splits so that they (and their contrasts) have joint inference.

Implementations

impl Apos

pub fn new( ml_g: impl Regressor + 'static, ml_m: impl Classifier + 'static, treatment_levels: Vec<f64>, ) -> Self

pub fn from_boxed( ml_g: Box<dyn Regressor>, ml_m: Box<dyn Classifier>, treatment_levels: Vec<f64>, ) -> Self

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

Propensity scores are clipped to [t, 1 − t] (default t = 0.01).

pub fn calibration(self, calibration: Calibration) -> Self

Calibrate the propensity scores before clipping (default: none), as DoubleML’s ps_processor_config.

pub fn normalize_ipw(self, normalize: bool) -> Self

Normalise the inverse-probability weights, as DoubleML’s normalize_ipw=True.

pub fn weights(self, weights: Array1<f64>) -> Self

Weighted average potential outcome, as DoubleML’s weights: E[w·g(d, X)] / E[w] (e.g. w = 1{X ∈ G} for a subgroup).

pub fn weights_with_bar( self, weights: Array1<f64>, weights_bar: Array2<f64>, ) -> Self

Weights with separate weights_bar for the inverse-probability correction, one column per repetition (DoubleML’s dict form).

pub fn n_folds(self, n_folds: usize) -> Self

Number of cross-fitting folds (default 5).

pub fn n_rep(self, n_rep: usize) -> Self

Number of independent sample splits (default 1).

pub fn seed(self, seed: u64) -> Self

Seed for the sample splits (default 0).

pub fn fit(&self, data: &DmlData) -> Result<AposFit>

Draw sample splits (stratified by the treatment, or by cluster for clustered data) and fit every level on them.

pub fn fit_with_folds( &self, data: &DmlData, splits: &[Vec<Fold>], ) -> Result<AposFit>

Fit every level on the given splits (one Vec<Fold> per repetition).

pub fn fit_with_cluster_folds( &self, data: &DmlData, splits: &[Vec<ClusterFold>], ) -> Result<AposFit>

Fit every level of clustered data on the given cluster splits.