joineR: jointSE – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

jointSE

Standard errors via bootstrap for a joint model fit

Description

This function takes a model fit from a joint model and calculates standard errors, with optional confidence intervals, for the main longitudinal and survival covariates.

Usage

jointSE(fitted, n.boot, gpt, lgpt, max.it, tol, print.detail = FALSE)

Arguments

`fitted`	a list containing as components the parameter estimates obtained by fitting a joint model along with the respective formulae for the longitudinal and survival sub-models and the model chosen, see `joint` for further details.
`n.boot`	argument specifying the number of bootstrap samples to use in order to obtain the standard error estimates and confidence intervals. Note that at least `n.boot = 100` is required in order for the function to return non-zero confidence intervals.
`gpt`	the number of quadrature points across which the integration with respect to the random effects will be performed. Defaults to `gpt = 3` which produces stable estimates in most datasets.
`lgpt`	the number of quadrature points which the log-likelihood is evaluated over following a model fit. This defaults to `lgpt = 10`, though `lgpt = 3` is often sufficient.
`max.it`	the maximum number of iterations of the EM algorithm that the function will perform. Defaults to `max.it = 200`, though more iterations may be necessary for large, complex data.
`tol`	the tolerance level before convergence of the algorithm is deemed to have occurred. Default value is `tol = 0.001`.
`print.detail`	This argument determines the level of printing that is done during the bootstrapping. If `TRUE` then the parameter estimates from each bootstrap sample are output.

Details

Standard errors and confidence intervals are obtained by repeated fitting of the requisite joint model to bootstrap samples of the original longitudinal and survival data. It is rare that more than 200 bootstrap samples are needed for estimating a standard error. The number of bootstrap samples needed for accurate confidence intervals can be as large as 1000.

Value

An object of class data.frame.

Author(s)

Ruwanthi Kolamunnage-Dona (ruwanthi.kolamunnage-dona@liverpool.ac.uk) and Pete Philipson (pete.philipson@northumbria.ac.uk)

References

Wulfsohn MS, Tsiatis AA. A joint model for survival and longitudinal data measured with error. Biometrics. 1997; 53(1): 330-339.

Efron B, Tibshirani R. An Introduction to the Bootstrap. 2000; Boca Raton, FL: Chapman & Hall/CRC.

Examples

data(heart.valve)
heart.surv <- UniqueVariables(heart.valve, 
                              var.col = c("fuyrs", "status"), 
                              id.col = "num")
heart.long <- heart.valve[, c("num", "time", "log.lvmi")]
heart.cov <- UniqueVariables(heart.valve, 
                             c("age", "hs", "sex"), 
                             id.col = "num")
heart.valve.jd <- jointdata(longitudinal = heart.long, 
                            baseline = heart.cov, 
                            survival = heart.surv, 
                            id.col = "num", 
                            time.col = "time")
fit <- joint(heart.valve.jd, 
             long.formula = log.lvmi ~ 1 + time + hs, 
             surv.formula = Surv(fuyrs, status) ~ hs, 
             model = "int")
jointSE(fitted = fit, n.boot = 1)

joineR

Joint Modelling of Repeated Measurements and Time-to-Event Data

v1.2.5

GPL-3 | file LICENSE

Authors

Pete Philipson [aut, cre] (<https://orcid.org/0000-0001-7846-0208>), Ines Sousa [aut] (<https://orcid.org/0000-0002-2712-1713>), Peter J. Diggle [aut] (<https://orcid.org/0000-0003-3521-5020>), Paula Williamson [aut] (<https://orcid.org/0000-0001-9802-6636>), Ruwanthi Kolamunnage-Dona [aut] (<https://orcid.org/0000-0003-3886-6208>), Robin Henderson [aut], Graeme L. Hickey [aut] (<https://orcid.org/0000-0002-4989-0054>), Maria Sudell [ctb], Medical Research Council [fnd] (Grant numbers: G0400615 and MR/M013227/1)

Initial release