unstructured – General structure variance models

Description

unstructured: General correlation and covariance models.

Usage

cor(obj, init=NA)
corv(obj, init=NA)
corh(obj, init=NA)
corb(obj, b=1, init=NA)
corbv(obj, init=NA)
corbh(obj, init=NA)
corg(obj, init=NA)
corgv(obj, init=NA)
corgh(obj, init=NA)
diag(obj, init=NA)
us(obj, init=NA)
chol(obj, k=1, init=NA)
ante(obj, k=1, init=NA)
sfa(obj, k=1, init=NA)
facv(obj, k=1, init=NA)
fa(obj, k=1, init=NA)
rr(obj, k=1, init=NA)

Arguments

`obj`	A factor in `data`.
`init`	A vector of initial values (correlation parameters followed by variance parameters) with an optional `names` attribute from the set {P, U, F} specifying the boundary constraint as positive, unconstrained or fixed, respectively.
`b`	Number of (sub-diagonal) bands in banded correlation models.
`k`	Order of the model (`chol, ante`) or number of factors (`sfa, facv, fa, rr`).

Details

The class of general variance models includes the simple, banded and general correlation models (cor, corb, corg), the diagonal, unstructured, Cholesky and antedependence variance models (diag, us, chol, cholc, ante) and the factor analytic structures (sfa, facv, fa).

Functions

`asr_corv:`	Simple correlation model, homogeneous variance form.
`asr_corh:`	Simple correlation model, heterogeneous variance form.
`asr_corb:`	Banded correlation model with `b`bands.
`asr_corbv:`	Banded correlation model with `b`bands, homogeneous variance form.
`asr_corbh:`	Banded correlation model with `b`bands, heterogeneous variance form.
`asr_corg:`	General correlation model.
`asr_corgv:`	General correlation model, homogeneous variance form.
`asr_corgh:`	General correlation model, heterogeneous variance form.
`asr_diag:`	Diagonal variance model.
`asr_us:`	Unstructured variance model.
`asr_chol:`	Cholesky variance model of order `k`.
`asr_ante:`	Antedependence variance model of order `k`.
`asr_sfa:`	Factor analytic model with `k` factors; the variance-covariance matrix is modelled on the correlation scale.
`asr_facv:`	Factor analytic model with `k` factors; the variance-covariance matrix is modelled on the covariance scale.
`asr_fa:`	Factor analytic model with `k` factors; sparse formulation where `k` “extra” levels are inserted in the mixed model equations.
`asr_rr:`	Factor analytic model with `k` factors; reduced rank formulation of `fa()` where the default boundary constraints for the specific variances are set to Fixed.

Updated on June 22, 2021

Was this article helpful?

Yes No