umxGSEM_std: Standardize GSEM S and V matrices to the correlation metric...

In umx: Structural Equation Modeling and Twin Modeling in R

Standardize GSEM S and V matrices to the correlation metric via Delta Method

Description

Standardizes the genetic covariance matrix S and propagates the uncertainty to the asymptotic sampling covariance matrix V using the Delta method. Standardizes the LDSC intercepts matrix I by dividing by the heritabilities.

Usage

umxGSEM_std(covstruc = NULL, S = NULL, V = NULL)

Arguments

covstruc	A list containing the genetic covariance matrix S, its asymptotic covariance V, and optional intercepts I.
S	Optional genetic covariance matrix.
V	Optional asymptotic covariance matrix.

Details

Standardizing S to a correlation matrix R and scaling V accordingly is highly recommended for model fitting to prevent numerical underflow and improve optimizer convergence rates when genetic variances are very small.

Value

A list containing the standardized covariance structure:

S	Standardized genetic correlation matrix.
V	Standardized asymptotic covariance matrix of the elements of S.
I	Standardized LDSC intercepts matrix (if present).

See Also

Other GSEM: Anthro_LDSC, Psych_LDSC, plot.umx_GWAS(), umxGSEM(), umxGSEM_GWAS(), umxGSEM_dl_tutorial_files(), umxGSEM_label_ldsc(), umxGSEM_ldsc(), umxGSEM_munge(), umxGSEM_sumstats(), umxSummary.MxDataStatic(), umxSummary.list()

Examples

# Example usage:
# std_cov = umxGSEM_std(covstruc)

umx documentation built on Oct. 11, 2026, 5:07 p.m.