Module select 

Fine–Hwang threshold estimators with pluggable statistics, plus an ensemble.

Every statistic produces, for lags 1..=p_n, a value and a null standard error. The estimator clips |stat_k| / se_k ≤ τ_n to zero and returns the last surviving lag (“last non-zero coefficient”, Fine & Hwang 1979, Def. 3).

Structs

EnsembleResult

FhConfig

Configuration of a single Fine–Hwang estimator.

FhResult

Output of fh_order.

JirakConfig

Configuration of jirak_order (Jirak 2012, q̂⁽⁵⁾).

JirakResult

Output of jirak_order.

Enums

EnsembleRule

How to combine several estimators.

Statistic

Which partial description Σ̂ of the process to threshold.

Functions

ensemble_order

Run several Fine–Hwang configurations and combine them.

fh_order

Fine–Hwang order estimator.

jirak_order

Jirak’s (2012, Ann. Statist. 40(2), eq. 2.14) refitted order estimator q̂⁽⁵⁾: fit Yule–Walker AR(k) for every dimension k ≤ d_n, studentize each coefficient vector with its exact null standard error se_i = sqrt(γ̂*_ii(k) · σ̂²(k) / n) (the Statistic::JirakBand studentization at dimension k), apply the common Gumbel threshold calibrated on the maximal dimension d_n, and return the largest per-dimension last-crossing.