dgCopula: Random generator for multivariate survival data via copulas.

In gsoutinho/survCopula: Multivariate Dependence Modeling with Copulas for Survival Data

Description

This function can be used to generate multivariate survival data in a variety of scenarios including competing risks, recurrent event and multi-state models.

Usage

dgCopula(
  typeCopula = "clayton",
  theta = 1,
  typeX = "Exp",
  num1_X = 1,
  num2_X = NULL,
  typeY = "Exp",
  num1_Y = 1,
  num2_Y = NULL,
  typeCens = "None",
  num1_Cens = NULL,
  num2_Cens = NULL,
  typeSurvData = "illness-death",
  state2.prob = 0.7,
  nsim = 250
)

Arguments

typeCopula	Type of copula. Possible options are "clayton", "frank" "FGM", "AMH", "gumbel-hougaard" and "joe". Defaults to "clayton".
theta	A numeric value for the space parameter.
typeX	Type of marginal distribution. Possible options are "Exp", "Norm" "Unif" and "Gamma". Defaults to "Exp".
num1_X	A numeric value for the first parameter of the first marginal distribution.
num2_X	A numeric value for the second parameter of the first marginal distribution. Only required for two parameter distributions.
typeY	Type of marginal distribution. Possible options are "Exp", "Norm", "Unif" and "Gamma". Defaults to "Exp".
num1_Y	A numeric value for the first parameter of the second marginal distribution.
num2_Y	A numeric value for the second parameter of the second marginal distribution. Only required for two parameter distributions.
typeCens	Type of censuring distribution. Possible options are "None", "Unif", "Exp" and "Wei". Defaults to "None".
num1_Cens	A numeric value for the first parameter of the censoring distribution.
num2_Cens	A numeric value for the second parameter of the censoring distribution. Only required for two parameter distributions.
typeSurvData	Type of survival data. Possible options are "time-to-event", "recurrent", #' "competing-risks" and "illness-death". Defaults to "illness-death".
state2.prob	Probability of a individual move to the intermediate state in illness-death model. Only required if typeSurvData =”illness-death”. Default to 0.7.
nsim	Number of observations to be generated.

Value

A numeric vector with the random multivariate survival data. Meaning of the colums for the type of survival data: Time-to-event:

T	Surival time. T = min(Y,Z).
Z	Censoring time variable.
Delta	Indicator status for censoring. Delta takes 1 when Y <= Z.

Recurrent:

T1	First gap time.
Delta1	Censoring indicator variable for the first gap time.
T2	Second gap time.
Delta2	Censoring indicator variable for the second gap time.
Z	Censoring time variable.

Competing risks:

T	Survival time.
Z	Censoring time variable.
Delta	Indicator status for censoring. Delta takes 0 if the competing risk process does not move from the initial state at the survival time T, or the value 1 and 2 for the possible causes of death 1 and 2.

Illness-death:

T1	Sojourn time in the initial state.
Delta1	Indicator status. Delta1 takes 1 when T1<T and T1<Z.
T	Total survival time.
Delta	Indicator status for censoring. Delta takes 1 when T<Z.
Z	Censoring time variable.

Author(s)

Gustavo Soutinho, Luis Meira-Machado

References

Meira-Machado, L.; de Una-Alvarez, J.; Cadarso-Suarez, C. and Andersen, P.K. Multi-state models for the analysis of time to event data. Statistical Methods in Medical Research, 2009, 18, 195-222.

Meira-Machado, L., Sestelo, M.; Gonlcalves, A. Nonparametric estimation of the survival function for ordered multivariate failure time data: A comparative study. Biometrical Journal, 2016, 58, 623-634.

L Meira-Machado, S Faria, A simulation study comparing modeling approaches in an illness-death multi-state model, Communications in Statistics-Simulation and Computation, 2014, 43 (5), 929-946

A Moreira, J de Una-Alvarez, L Machado Presmoothing the Aalen-Johansen estimator in the illness-death model, Electronic Journal of Statistics, 2013, 7, 1491-1516

A Moreira, L Meira-Machado, survivalBIV: Estimation of the bivariate distribution function for sequentially ordered events under univariate censoring, J Stat Softw, 2012, 46 (13), 1-16

Examples

sim.data<-dgCopula(typeCopula ='clayton', theta=1, typeX='Unif', num1_X=0, num2_X=5, 
                  typeY='Unif',  num1_Y=0, num2_Y=5,  typeCens='Unif', num1_Cens=0, 
                  num2_Cens=7,  nsim=250,typeSurvData='time-to-event')

head(sim.data)
            
sim.data2<-dgCopula(typeCopula ='frank', theta=10, typeX='Exp', num1_X=0.5, 
                   typeY='Exp', num1_Y=1.5, nsim=250,typeSurvData='illness-death', 
                   typeCens='Unif', num1_Cens=0, 
                   num2_Cens=4, state2.prob=0.6)
            

gsoutinho/survCopula documentation built on Sept. 4, 2020, 3:54 a.m.