jmdl: Fit Joint Mean-Correlation Models For Discrete Longitudinal...

In jmdl: Joint Mean-Correlation Regression Approach for Discrete Longitudinal Data

Description

Fit a joint mean-correlation model to discrete longitudinal data.

Usage

jmdl(formula, data = NULL, q = 2, theta = NULL, W.appendix = NULL,
  offset = NULL, family = c("Bernoulli", "Nbinom", "Poisson"))

Arguments

formula	a two-sided linear formula object describing the correlation for both the mean and correlation matrix part of the model, with the response, the corresponding subject id and measurement time on the left of a operator~, divided by vertical bars ("|").
data	data frame containing the variables named in formula.
q	degree of polynomial of the time lag to model the lower triangular matrix.
theta	starting values for the parameters in the model.
W.appendix	appendix array to model time-dependent covariates for the lower triangular matrix.
offset	a term to be added to the linear predictor.
family	the marginal distributions of the discrete variables. choose 'Bernoulli', 'Poisson' or 'Nbinom'.

Examples

data(toydata)
mydat <- toydata
fit <- jmdl(Y|id|time ~ X, data = mydat, q = 2, family ='Bernoulli')

jmdl documentation built on May 2, 2019, 11:04 a.m.