Return a 2 dimensional array, with subjects in the first dimension and transition (numbers) in the second. Can expand later to include time-dependent covariates by introducing extra dimension. Entries of the matrix are exp(lp)_i^m, with i denoting the subject and m the transition.
trans_specific_risks(object, newdata, trans)A coxph object fit on multi-state
data. Must contain a strata(X) term. Data used for the coxph() fit preferably prepared
using msprep.
A data.frame containing a single row for each transition per
subject in the data. For a model with m possible transitions, and n subjects
newdata must have m*n rows. It must contain the following named columns:
id:Unique identifier of the subject, can be numeric or character;
from:State from which the transition takes place;
to:State to which the transition takes place;
trans:Transition number in the 'transMat' trans this transition relates to;
"variables":The variables and their values for the subject
identified by "id" for the transition this entry relates to. Names must
match the names of the variables in coxph object.
Note that newdata must contain a column containing the variable which was
used to determine the stratum of a transition in object.
Usually the stratum is determined from one of the required columns. The
"variables" columns can usually be obtained using expand.covs.
A transition matrix as created by transMat.