acc (version 1.3.3)

aeexfit: Fits semiparametric regression models robust to informative observation times and censoring

Description

Fits semiparametric regression models robust to informative observation times and censoring

Usage

aeexfit(formula, data, weight, se, control=list(), boot)

Arguments

formula

A formula object as returned by aee.

data

A data frame which includes individuals' ID, observation times, and minutes of physical activity since the last observation time.

weight

A vector of sampling weights, for each individual.By default, no sampling weights are applied.

se

The method of estimating standard errors can be chosen by the argument se. Two options are available: i) the sandwhich estimation (se = 'Sandwich'), or ii) the bootstrap procedure (se = 'Bootstrap').

control

A list of control parameters. See 'Details'.

boot

The number of resamples generated for the bootstrap procedure.

Details

The control argument is a list athat can supply any of the following components:

• betaInit: Initial value for covariate coefficient, default is 0.

• interval: Initial search interval for solving beta. Default is (-5,5).

• maxIter: Maximum iterations allowed. Default is 150.

• absTol: Absolute tolerance. Default is 1e-6.

• relTol: Relative tolerance. Default is 1e-6.

• a: A tune parameter. Default is .1. In case of gamma fraility, "a" corresponds to the value of both shape and rate parameters.

References

Wang, X. and Yan, J. (2011). Fitting semiparametric regressions for panel count survival data with an R package spef. Computer Methods and Programs in Biomedicine, 104, 278-285.

Wang, X., Ma, S., and Yan, J. (2013). Augmented estimating equations for semiparametric panel count regression with informative observation times and censoring time. Statistica Sinica, 23, 359-381.

Examples

## Not run: 
# 
# data(NHANES)
# formula <- aee(ID, Day, mvpaMinutes) ~ VO2max
# # Standard errors are obtained using sandwich estimation
# fitted <- aeexfit(formula = formula, data = NHANES, se = "Sandwich")
# summary(fitted)
# ## End(Not run)