Cprob (version 1.4.1)

cpfpo: Proportional-odds Model for the Conditional Probability Function

Description

This function uses the Temporal Process Regression framework to fit a proportional-odds model to the conditional probability function.

Usage

cpfpo(formula, data, subset, na.action, failcode, tis, w, ...)

Arguments

formula

A formula object whose response, on the left of a ~ operator, is a Hist object, and the terms on the right of ~

data

A data.frame in which to interpret the variable names in the formula and subset

subset

Expression specifying that only a subset of the data set should be used

na.action

A missing data filter funtion applied to the model.frame, after any subset argument has been used. Default is options()$na.action

failcode

Integer specifying the code for the event of interest

tis

Vector of timepoints on which the model is fitted

w

Vector of weights. Should be of the same length as tis. Default is rep(1, length(tis))

…

Further arguments for tpr

Value

cpfpo returns an object of class cpfpo and tpr. See tpr for further details.

Details

The conditional probability function of a competing event is the probability of having failed due to one risk (the event of interest) given that no other failure has previously occurred.

The cpfpo function fits a proportional-odds model for the conditional probability function within the Temporal Process Regression framework, which is a marginal mean model, where the mean of a response \(Y(t)\) at time \(t\) is specified conditionally on a vector of covariates \(Z\) and a time-dependent stratification factor \(S(t)\) $$E\{Y(t) | Z, S(t) = 1\} = g^{-1}\{\beta(t)'Z\}$$ This approach enables the application of standard binary regression models in continuous time.

The regression model is fitted using the tpr package. See tpr for further details.

References

J.P. Fine, J. Yan and M.R. Kosorok (2004). Temporal Process Regression, Biometrika, 91(3):683-703.

M.S. Pepe and M. Mori, Kaplan-Meier, marginal or conditional probability curves in summarizing competing risks failure time data? Statistics in Medicine, 12(8):737--751.

A. Allignol, A. Latouche, J. Yan and J.P. Fine (2011). A regression model for the conditional probability of a competing event: application to monoclonal gammopathy of unknown significance. Journal of the Royal Statistical Society: Series C, 60(1):135--142.

See Also

tpr, print.cpfpo, xyplot.cpfpo

Examples

# NOT RUN {
data(mgus)
mgus$A <- ifelse(mgus$age < 64, 0, 1)

## fit the model for 2 covariates
fit.cpfpo <- cpfpo(Hist(time, ev)~factor(A) + creat,
                   data = mgus, tis=seq(10, 30, 0.3),
                   w=rep(1,67))
## and plot the odds-ratios
if(require("lattice")) {
xyplot(fit.cpfpo, scales = list(relation = "free"), layout = c(3, 1))
}
# }