Struct FrontdoorCriterion 

pub struct FrontdoorCriterion {
    pub graph: CausalGraph,
}

Implements the frontdoor adjustment formula for causal effect identification.

The frontdoor criterion allows identification of causal effects through a mediator set M when direct adjustment is not possible.

Fields

graph: CausalGraph

The causal graph.

Implementations

impl FrontdoorCriterion

pub fn new(graph: CausalGraph) -> Self

Create a new frontdoor criterion object.

pub fn check( &self, treatment: usize, outcome: usize, mediator_set: &[usize], ) -> bool

Check if mediator_set satisfies the frontdoor criterion for identifying the effect of treatment on outcome.

Conditions:

1. All directed paths from treatment to outcome are intercepted by M.
2. No backdoor path from treatment to M (or blocked by treatment).
3. All backdoor paths from M to outcome are blocked by treatment.

pub fn adjusted_effect( data_x: &[f64], data_m: &[f64], data_y: &[f64], x_val: f64, ) -> f64

Compute the frontdoor-adjusted causal effect using sample data.

E[Y|do(X=x)] = Σ_m P(M=m|X=x) Σ_x’ E[Y|M=m, X=x’] P(X=x’)

Simplified discrete approximation for a single mediator.

Trait Implementations

impl Clone for FrontdoorCriterion

fn clone(&self) -> FrontdoorCriterion

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for FrontdoorCriterion

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

Formats the value using the given formatter.