reReg (version 1.3.0)

simSC: Function to generate simulated recurrent event data

Description

The function simSC() generates simulated recurrent event data from either a Cox-type model, an accelerated mean model, an accelerated rate model, or a scale-change model.

Usage

simSC(
  n,
  a1 = a2,
  b1 = b2,
  a2 = a1,
  b2 = b1,
  type = "cox",
  zVar = 0.25,
  tau = 60,
  summary = FALSE
)

Arguments

n

number of observation.

a1, a2, b1, b2

are numeric vectors of length 2. These correspond to the $\alpha$, $\beta$, $\eta$, and $\theta$ in the joint model. See Details

type

is a character string specifying the underlying model. The rate function type and the hazard function type are separated by a vertical bar "|", with the rate function on the left. For example, type = "cox|am" generates the recurrent process from a Cox model and the terminal event from an accelerated mean model. Setting type = "cox" gives type = "cox|cox".

zVar

a numeric variable specifying the variance of the fraility variable,$Z$, when zVar > 0. When zVar = 0, $Z$ is set to a fixed constant 1. The default value is 0.25.

tau

a numeric value specifying the maximum observation time.

summary

a logical value indicating whether a brief data summary will be printed.

Details

The function simSC() generates simulated recurrent event data over the interval $(0,\tau )$ based on the specification of the recurrent process and the terminal events. Specifically, the rate function, $\lambda (t)$, of the recurrent process can be specified as one of the following model: $$\lambda (t)=Z{\lambda }_0(t{e}^{X^{\mathrm{\top }}\alpha })e^{X^{\mathrm{\top }}\beta },h(t)=Z{h}_0(t{e}^{X^{\mathrm{\top }}\eta })e^{X^{\mathrm{\top }}\theta },$$ where ${\lambda }_0(t)$ is the baseline rate function, $h_0(t)$ is the baseline hazard function, $X$ is a $n$ by $p$ covariate matrix and $\alpha$, $Z$ is an unobserved shared frailty variable, and $(\alpha ,\eta )$ and $(\beta ,\theta )$ correspond to the shape and size parameters of the rate function and the hazard function, respectively.

For all scenarios, two covariates are considered; $X=(X_1,X_2)$, where $X_1$ follows a Bernoulli distribution with probability 0.5 and $X_2$ follows a standard normal distribution. The censoring time could be either independent (given covariates) or informative. The simulated data is used for illustration. An informative censoring time, $C$, is generated separately from an exponential distribution with a rate parameter of 1 / 60 if $X_1$ is 1, or $Z^2/30$ if $X_1$ is 0. The observed recurrent events is then observed up to the minimum of $C$, terminal event, and $\tau$. Lastly, we assume the baseline functions $${\lambda }_0(t)=\frac{2}{1+t},h_0(t)=\frac{1}{8(1+t)}.$$

See Also

reReg

Examples

# NOT RUN {
set.seed(123)
simSC(100, c(-1, 1), c(-1, 1), summary = TRUE)
# }