Grow random survival tree using multivariate longitudinal endogenous covariates
DynTree(
Y,
Longitudinal = NULL,
Numeric = NULL,
Factor = NULL,
timeVar = NULL,
mtry = 1,
nsplit_option = "quantile",
nodesize = 1,
seed = 1234
)A list of output which should contain: type defines the nature of the outcome, can be "surv", "numeric" or "factor"; Y is the output variable; id is the vector of the identifiers for each individuals, they should be the same as the identifiers of the inputs.
A list of longitudinal predictors which should contain: X a dataframe with one row for repeated measurement and as many columns as markers; id is the vector of the identifiers for the repeated measurements contained in X; time is the vector of the measurement times contained in X.
A list of numeric predictors which should contain: X a dataframe with as many columns as numeric predictors; id is the vector of the identifiers for each individual.
A list of factor predictors which should contain: X a dataframe with as many columns as factor predictors; id is the vector of the identifiers for each individual.
A character indicating the name of time variable
Number of candidate variables randomly drawn at each node of the trees. This parameter should be tuned by minimizing the OOB error. Default is NULL.
A character indicates how the values are chosen to build the two groups for the splitting rule (only for continuous predictors). Values are chosen using deciles (nsplit_option="quantile") or randomly (nsplit_option="sample"). Default value is "quantile".
Minimal number of subjects required in both child nodes to split. Cannot be smaller than 1.
Seed to replicate results