Struct JumpDiffusionPriorConfig 

#[non_exhaustive]
pub struct JumpDiffusionPriorConfig {
    pub detection: JumpEventConfig,
    pub min_jump_count: usize,
    pub min_input_observations: usize,
    pub diffusion_correction: bool,
    pub noise_scale: JumpPriorNoiseScale,
    pub max_jump_fraction: f64,
    pub truncation_correction: bool,
    pub min_detection_probability: f64,
}

Configuration for jump_diffusion_prior_signal.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

detection: JumpEventConfig

Detection settings shared with jump_event_signal. Detection depth defaults to scale level 1 — and, unlike jump_event_signal, a None levels does not inherit the profile’s multi-scale max_levels: a Merton jump is a one-period impulse, the level-1 MODWT statistic is the one with an exactly calibrated null, and deeper scales would misattribute multi-day volatility bursts to the jump intensity, so depth beyond 1 is strictly opt-in. near_threshold settings are ignored here (sub-threshold scores never enter the triple).

min_jump_count: usize

Minimum consolidated jumps required for an identifiable triple. Below this the signal is Unavailable — never a fabricated estimate. Must be at least 2 (a jump-size std needs two observations).

min_input_observations: usize

Minimum finite price-return observations required. The effective floor is the max of this and detection.min_input_observations.

diffusion_correction: bool

Subtract the per-period bipower diffusive variance from the raw jump-size variance before taking σ_J. A detected jump return is jump + diffusion, so raw moments overstate σ_J by the diffusive floor; the correction also aligns prior_jump_variance with the realized QV − BV leg, which subtracts the same floor.

noise_scale: JumpPriorNoiseScale

Noise scale for the detection test statistic. See JumpPriorNoiseScale; defaults to JumpPriorNoiseScale::Mad.

max_jump_fraction: f64

Maximum fraction of the window’s returns that may carry a consolidated detection before the estimate is refused (implausible_jump_fraction). Merton jumps are rare by model assumption: when the detector flags this much of the window, the Gaussian-null noise scale itself is not credible (e.g. a near-degenerate coefficient distribution collapsing the MAD to rounding residue) and any triple would be fabricated. Default 0.2 — beyond ~50 jumps/year at daily resolution the jump/diffusion separation is meaningless anyway. Must be in (0, 1]. The cap applies to detected jumps; a truncation-corrected λ may imply a total fraction above it (reported via diagnostic, not re-gated).

truncation_correction: bool

Correct the detection-conditional λ for threshold truncation (#86): fit the jump-size distribution as an outer-truncated normal (EM), compute the detection probability p̂_det implied by the fitted distribution and the level-1 cutoff, and inflate λ̂ by 1/p̂_det (widening its SE through the correction). Applies only on the default detector shape — MODWT, level-1, MAD noise scale; anywhere else, or on any numerical failure, the uncorrected detection-conditional triple is published with a truncation_correction_skipped diagnostic. Default true.

min_detection_probability: f64

Floor on the fitted detection probability below which the truncation correction is not applied (the window cannot identify total intensity; inflating by 1/p̂ would be fabrication). Default 0.2; must be finite and in (0, 1).

Implementations

impl JumpDiffusionPriorConfig

pub fn for_daily_eod() -> Self

Default prior-estimation settings for daily EOD analytics.

pub fn validate(&self) -> Result<()>

Validate the prior-estimation configuration.

Trait Implementations

impl Clone for JumpDiffusionPriorConfig

fn clone(&self) -> JumpDiffusionPriorConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for JumpDiffusionPriorConfig

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

Formats the value using the given formatter.

impl Default for JumpDiffusionPriorConfig

fn default() -> Self

Returns the “default value” for a type.

impl PartialEq for JumpDiffusionPriorConfig

fn eq(&self, other: &JumpDiffusionPriorConfig) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for JumpDiffusionPriorConfig