errorVar function

Error scale matrix associated with times

It returns a scale matrix associated with the error term at time times. Can be applied to a smn.lmm or smsn.lmm object or to a specific dependence structure with chosen parameter values.

errorVar(times, object = NULL, sigma2 = NULL, depStruct = NULL,
    phi = NULL)

Arguments

• times: A vector containing the times for which the matrix should be calculated.
• object: A smn.lmm or smsn.lmm object for which the variance should be extracted.
• sigma2: Common variance parameter, such that $\Sigma = sigma2*R$. Only evaluated if object is.null(object).
• depStruct: Dependence structure. "UNC" for conditionally uncorrelated ("CI" is also accepted), "ARp" for AR(p) -- p is length(phi)--, "CS" for compound symmetry, "DEC" for DEC, and "CAR1" for continuous-time AR(1). Only evaluated if object is.null(object).
• phi: Parameter vector indexing the dependence structure. Only evaluated if object is.null(object).

Returns

Matrix of dimension length(times).

References

Schumacher, F. L., Lachos, V. H., and Matos, L. A. (2021). Scale mixture of skew-normal linear mixed models with within-subject serial dependence. Statistics in Medicine 40(7), 1790-1810.

Author(s)

Fernanda L. Schumacher, Larissa A. Matos and Victor H. Lachos

See Also

smsn.lmm, smn.lmm

Examples

fm1 = smsn.lmm(distance ~ age+Sex, data=nlme::Orthodont, groupVar="Subject")
errorVar(times=1:4,fm1)
#
errorVar(times=1:5, sigma2=1, depStruct="ARp", phi=.5)
errorVar(times=1:5, sigma2=1, depStruct="DEC", phi=c(.5,.8))