dot-add_leverage: Add leverage estimation to a fitted SV(p) model...

In wARMASVp: Winsorized ARMA Estimation for Higher-Order Stochastic Volatility Models

Add leverage estimation to a fitted SV(p) model (post-processing) Works for all error distributions: Gaussian, Student-t, GED. - Gaussian: closed form (C_F = 1) - Student-t: closed form with C_t(nu) correction - GED: exact 1D root-finding via uniroot + Gauss-Hermite quadrature

Description

Add leverage estimation to a fitted SV(p) model (post-processing) Works for all error distributions: Gaussian, Student-t, GED. - Gaussian: closed form (C_F = 1) - Student-t: closed form with C_t(nu) correction - GED: exact 1D root-finding via uniroot + Gauss-Hermite quadrature

Usage

.add_leverage(out, y, p, rho_type, del, trunc_lev, wDecay, errorType)

Arguments

out	Model object from .svp_gaussian/.svp_t/.svp_ged (without leverage)
y	Numeric vector of observations
p	AR order
rho_type	"pearson", "kendall", or "both"
del	Small constant for log transform
trunc_lev	Logical: truncate rho to [-0.999, 0.999]
wDecay	Logical: decaying weights (passed from original estimation)
errorType	"Gaussian", "Student-t", or "GED"

Value

Updated model object with leverage fields added

wARMASVp documentation built on May 15, 2026, 5:07 p.m.