AICcond: Computation of the Conditional Akaike Information Criterion...

In VivianePhilipps/multLPM: Joint model for multivariate latent processes and clinical endpoints using an exact likelihood inference method

Description

Computation of the Conditional Akaike Information Criterion (AICcond) for a joint model estimated by JointMult function

Usage

AICcond(model, Y, D, data, var.time, RE = "block-diag", BM = "diag", B,
  posfix, breaks = NULL, delayed = TRUE)

Arguments

model	a JointMult model
Y	a list of multlcmm objects
D	a list of two-sided formula defining the event part of the model
data	data.frame containing the observations and variables
var.time	a character vector indicating the name of the time variables
RE	an indicator of the random effect structure between dimensions
BM	an indicator of the correlation of the Brownian motions
B	vector containing initial values for the parameters
posfix	optional vector specifying the indices in vector B of the parameters that are not estimated
breaks	optional vector specifying the break points in the case where the event time is discretized
delayed	logical vector indicating if delayed entry should be accounted for

Value

A list containing :

AICcond	the conditional Akaike Information Criterion
vrais2i	a vector containing individual contributions to the conditional and total log-likelihood
npm	the number of estimated parameters for the joint model
npmtot	a vector containing the number of estimated parameters of each longitudinal submodel

Author(s)

Tiphaine Saulnier, Cecile Proust-Lima and Viviane Philipps

References

Zhang et al, Assessing model fit in joint models of longitudinal and survival data with applications to cancer clinical trials. Statistics in medicine 2014 vol. 33, no 27, p. 4715-4733.

VivianePhilipps/multLPM documentation built on March 13, 2021, 6:35 a.m.