jmo_0: Joint Modelling for Ordinal outcomes

Description

Joint modeling of longitudinal ordinal data and competing risks

Usage

jmo_0(
  p,
  s,
  yfile,
  cfile,
  mfile,
  point = 20,
  maxiterations = 1e+05,
  do.trace = FALSE,
  type_file = TRUE
)

Arguments

The dimension of proportional odds covariates (not including intercept) in yfile.

The dimension of non-proportional odds covariates in yfile.

yfile

Y matrix for longitudinal measurements in long format. For example, for a subject with n measurements, there are n rows for this subject. The # of rows in y matrix is the total number of measurements for all subjects. The columns in Y are ordered this way: the longitudinal outcome (column 1), then the covariates for random effects, and lastly, the covariates for fixed effects (no intercept).

cfile

C matrix for competing risks failure time data. Each subject has one data entry, so the number of rows equals to the number of subjects. The survival / censoring time is included in the first column, and the failure type coded as 0 (censored events), 1 (risk 1), or 2 (risk 2) is given in the second column. Two competing risks are assumed. The covariates are included in the third column and on.

mfile

M vector to indicate the number of longitudinal measurements per subject. The number of rows equals to the number of subjects.

point

Quadrature points used in the EM procedure. Default is 20.

maxiterations

Maximum values of iterations. Default is 100000.

do.trace

Print detailed information of each iteration. Default is false, not to print the iteration details.

type_file

Types of inputs. Default is true, i.e. data files with headers. If set to "F", inputs are changed to data matrixes or data.frames (with headers)

Value

Object of class JMcmprsk with elements

`vcmatrix`	The variance-covariance matrix for all the parameters. The parameters are in the order: \(\beta\), \(\alpha\), \(\theta\), \(\gamma\), \(\nu\), and \(\Sigma\). The elements in \(\Sigma\) are output in the order along the main diagonal line, then the second main diagonal line, and so on.
`betas`	The point estimates of \(\beta\).
`se_betas`	The standard error estimate of \(\beta\).
`alphamatrix`	The point estimates of \(\alpha\).
`se_alphas`	The standard error estimate of \(\alpha\).
`theta`	The point estimates of \(\theta\).
`se_theta`	The standard error estimate of \(\theta\).
`gamma_matrix`	The point estimate of \(\gamma\).
`se_gamma_matrix`	The standard error estimate of \(\gamma\).
`v_estimate`	The point estimate of \(\nu\).
`se_v_estimate`	The standard error estimate of \(\nu\).
`sigma_matrix`	The point estimate of \(\Sigma\) (only the upper triangle portion of the matrix is output).
`se_sigma`	The standard error estimate of \(\Sigma\).The standard errors are given in this order: main diagonal, the second main diagonal, and so on.
`loglike`	Log Likelihood.

References

Ning Li,Robert M. Elashoff,Gang Li and Jeffrey Saver. "Joint modeling of longitudinal ordinal data and competing risks survival times and analysis of the NINDS rt-PA stroke trial." Statistics in medicine 29.5 (2010): 546-557.

Examples

Run this code

# NOT RUN {
require(JMcmprsk)
set.seed(123)

# A toy example on a dataset called from file paths
yfn=system.file("extdata", "jmosimy.txt", package = "JMcmprsk")
cfn=system.file("extdata", "jmosimc.txt", package = "JMcmprsk")
mfn=system.file("extdata", "jmosimm.txt", package = "JMcmprsk")
fit <- jmo_0(p=3,s=1, yfn,cfn,mfn,point=6,do.trace = FALSE)
fit

# }
# NOT RUN {
# A toy example on a dataset called from data frame
data(ninds)
yread <- ninds[, c(2:14)]
mread <- as.data.frame(table(ninds$ID))
mread <- as.data.frame(mread[, 2])
cread <- ninds[, c(1, 15, 16, 6, 10:14)]
cread <- unique(cread)
cread <- cread[, -1]
jmofit=jmo_0(p=9,s=2, yread,cread,mread,point=6,do.trace = FALSE, type_file = FALSE)
jmofit
# }