This variance function accepts a list of matrices, which define a correlation structure for a coxme fit.
coxmeMlist(varlist, rescale = FALSE, pdcheck = TRUE, positive = TRUE)a list containing one or more matrix or bdsmatrix objects.
if TRUE, each input matrix is rescaled to have a diagonal of 1. (Kinship matrices for instance are often generated with a diagonal of .5 and would be multiplied by 2).
check each matrix to ensure that it is positive definite
constrain coefficients to be positive. This may also be a vector of
the same length as varlist
a coxme variance family object, used by coxme in the fitting process.
If two matrices \(A\) and \(B\) were given, this fits
the variance structure
\(V= \sigma_1^2 A + \sigma_2^2 B\),
where the variances \(\sigma_1^2\) and \(\sigma_2^2\) are
parameters that will be optimized by coxme,
treating \(A\) and \(B\) as fixed.