Struct PriorSensitivity 

pub struct PriorSensitivity {
    pub scales: Arc<[f64]>,
    pub alphas: Arc<[f64]>,
    pub max_relative_range: f64,
}

Prior sensitivity grid: isotropic scales or external α multipliers.

Fields

scales: Arc<[f64]>

Prior scales (σ of isotropic Gaussian coefficient prior). Empty in α mode.

alphas: Arc<[f64]>

Multipliers on post-conflict applied alphas. Empty in isotropic scale mode.

max_relative_range: f64

Fail when (max−min) / scale exceeds this, where scale is max(|means…|, |original_ate|, ε).

Implementations

impl PriorSensitivity

pub fn standard_grid() -> Self

Standard isotropic grid {0.5, 1, 2, 5, 10, 20} with DEFAULT_MAX_RELATIVE_PRIOR_RANGE.

pub fn standard_alpha_grid() -> Self

Standard external-α multiplier grid {0, 0.25, 0.5, 0.75, 1}.

Multiplier 0 is baseline-only; 1 uses full post-conflict applied alphas.

pub fn evaluate( &self, estimator: &BayesianGComputationAte, problem: &PreparedBayesianProblem, identification: IdentificationStatus, workspace: &mut BayesianGCompWorkspace, ctx: &ExecutionContext, ) -> Result<(PriorSensitivitySummary, Vec<CausalPosterior>), ValidationError>

Refit Bayesian g-comp at each prior scale; return sensitivity summary.

Errors

Fit failures or empty scale grid.

pub fn evaluate_external_alpha( &self, estimator: &BayesianGComputationAte, problem: &PreparedBayesianProblem, identification: IdentificationStatus, workspace: &mut BayesianGCompWorkspace, ctx: &ExecutionContext, external: ExternalAlphaSensitivity<'_>, ) -> Result<(PriorSensitivitySummary, Vec<CausalPosterior>), ValidationError>

Refit at each α-multiplier on post-conflict applied alphas (external prior bank).

For multiplier m, composed alphas are m * alphas_applied[k] (clamped to [0, 1]).

Errors

Empty α grid, length mismatch, compose failures, or fit failures.

pub fn to_report( &self, summary: &PriorSensitivitySummary, original_ate: f64, ) -> RefutationReport

Convert sensitivity range into a refutation-style report.

Passes when the relative range of effect means is finite and ≤ max_relative_range.

Trait Implementations

impl Clone for PriorSensitivity

fn clone(&self) -> PriorSensitivity

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PriorSensitivity

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

Formats the value using the given formatter.

impl Default for PriorSensitivity

fn default() -> Self

Returns the “default value” for a type.