parametrize function

Parametrizations of Covariates and Covariate Coefficients

This function helps the parametrizations of covariates and covariate coeffcients when users specify a general hazard rate function in function simEvent and simEventData. It applies the specified function (or the built-in option) FUN to the $i_{th}$ row of the covariate matrix z and the $i_{th}$ row of the coefficient matrix, iteratively, for $i$ from one to the number of rows of the covariate matrix z.

parametrize(z, zCoef, FUN = c("exponential", "linear", "excess"), ...)

Arguments

• z: A numeric matrix, each row of which represents the covariate vector at one perticular time point.
• zCoef: A numeric matrix, each row of which represents the covariate coeffcient vector at one perticular time point.
• FUN: The parametrization of the model parameter(s) with covariates and covariate coefficients. The built-in options include "exponential", "linear", "excess" for parametrization in the exponential, linear, excess relative risk model form, respectively. It can also be a function that at least has argument z and zCoef for incorporating the covariates and covariate coefficients into the model. The user-specified function should expect that both the input z and zCoef are numeric vectors and return a numeric value (or can be convected to a numeric value by as.numeric).
• ...: Other arguments that can be passed to the function FUN.

Returns

A numeric vector.

Examples

## time points
timeVec <- c(0.5, 2)
## time-variant covariates
zMat <- cbind(0.5, ifelse(timeVec > 1, 1, 0))
## time-varying coefficients
zCoefMat <- cbind(sin(timeVec), timeVec)

## the following three ways are equivalent for the exponential form,
## where the first one (using the built-in option) has the best performance
parametrize(zMat, zCoefMat, FUN = "exponential")
parametrize(zMat, zCoefMat, function(z, zCoef) exp(z %*% zCoef))
sapply(1 : 2, function(i) as.numeric(exp(zMat[i, ] %*% zCoefMat[i, ])))

See Also

simEvent