Cox Proportional Hazards Dialog Function

DESCRIPTION:

These functions are used by the Cox Proportional Hazards dialog. menuCoxph calls tabSummary.coxph, tabPredict.coxph and tabSurvfit.coxph if summary, prediction, and survival curve results are requested.

USAGE:

menuCoxph(formula, data, subset=NULL, na.omit.p, weights, eps, init,  
          iter.max, method, singular.ok, robust, print.short=F,  
          print.long=T, predict.p=F, newdata=NULL, save.name=NULL,  
          type="lp", se.fit=F, collapse=NULL, plot.fit=F, newdata2=NULL,  
          individual=F, plot.conf.int=T, col, lty, lwd, log, yscale,  
          xscale, xlab, ylab, xaxs, curve.type, conf.int, conf.type) 
tabSummary.coxph(fit, print.short, print.long) 
tabPredict.coxph(object, newdata = NULL, save.name, type = "lp", se.fit = F, 
          collapse = NULL, predict.p=T) 
tabSurvfit.coxph(object, plot.fit, newdata, individual, type, plot.conf.int,  
          conf.int, conf.type, se.fit, col, lty, lwd, log, yscale, xscale,  
          xlab, ylab, xaxs) 

REQUIRED ARGUMENTS:

formula
a formula object having a Surv object as the response on the left of a ~ operator, and the terms, separated by + operators, on the right.
fit
an object of class coxph.
object
an object of class coxph.

OPTIONAL ARGUMENTS:

data
an optional data frame in which to interpret the variables occurring in the formula.
subset
an expression saying which subset of the rows of the data should be used in the fit. This can be a logical vector (which is replicated to have length equal to the number of observations), or a numeric vector indicating which observation numbers are to be included, or a character vector of the row names to be included. All observations are included by default.
na.omit.p
if TRUE, then any observation with missing values are removed from the analysis. If FALSE and there are missing values then the function will exit with a message that missing values are not allowed. If na.omit.p is TRUE then na.action is set to na.omit in the call to coxph. If na.omit.p is FALSE then na.action is set to na.fail in the call to coxph.
weights
the optional case weights.
eps
convergence threshold. Iteration will continue until the relative change in the log-likelihood is less than eps.
init
vector of initial values of the iteration. In the call to coxph, the initial value for all variables is 0, by default.
iter.max
maximum number of iterations to perform.
method
a character string specifying the method for tie handling. Possible values are "efron", "breslow", and "exact".
singular.ok
if TRUE, columns of the X matrix that are linear combinations of earlier columns will be skipped.
robust
if TRUE, a robust variance estimate is returned.
print.short
if TRUE, a short summary of the nonparametric survival model is printed. This output is from the function print.coxph.
print.long
if TRUE, a long summary of the nonparametric survival model is printed. This output is from the function summary.coxph.
predict.p
if TRUE, predictions will be saved to the data frame specified in save.name.
newdata
a data frame to use for computing predictions. It must contain the same names as those used in the original analysis. If missing, the predictions for the original data are computed.
save.name
a character string for the name of the data frame to save the predictions and standard errors in. If a data frame with this name already exists in database 1 and it has the appropriate number of rows then the saved values will be appended to the data frame. If the object already exist in database 1 and it is not a data frame or it does not have the appropriate number of rows then a new name is created by appending a number to save.name and the results are saved in the data frame with the new name.
type
a character string indicating the type of predictions. Possible values are "lp", "risk", "expected", and "terms".
se.fit
if TRUE, standard errors will be saved to the data frame specified in save.name.
collapse
a grouping variable over which to collapse the predictions. predict.coxph will sum the predictions for each level of this variable.
plot.fit
if TRUE, a plot of the survival curves is created.
newdata2
a data frame with the same variable names as those that appear in formula. The value of newdata2 is passed to the newdata argument in tabSurvfit.coxph.
individual
if TRUE, the data frame represents different time epochs for only one individual. If FALSE, multiple rows indicate multiple individuals.
plot.conf.int
if TRUE, confidence intervals are included on the plot.
col
a list of integers specifying colors for each curve. The list is converted to a vector for the call to survfit. Elements of the list are reused cyclically if the length of the list is shorter than the number of curves.
lty
a list of integers specifying line types for each curve. The list is converted to a vector for the call to survfit. Elements of the list are reused cyclically if the length of the list is shorter than the number of curves.
lwd
a list of numeric values for line widths for each curve. The list is converted to a vector for the call to survfit. Elements of the list are reused cyclically if the length of the list is shorter than the number of curves.
log
if TRUE, the y-axis will be on the log scale.
yscale
a numeric value used to multiply the labels on the y-axis.
xscale
a numeric value used like yscale for labels on the x-axis.
xlab
a character string used to label the x-axis.
ylab
a character string used to label the y-axis.
xaxs
a character string denoting the x-axis style. Possible values are "standard", "tight", and "extended". These values are converted to NULL, "i", and "e", respectively, for the call to plot.survfit.
curve.type
a character string specifying the type of survival curve. Possible values are "aalen" and "kaplan-meier".
conf.int
the level for a two-sided confidence interval on the survival curve(s).
conf.type
a character string specifying the confidence interval type. Possible values are "none" for no confidence intervals, "plain" for standard intervals, "log" for intervals based on the cumulative hazard, and "log-log" for intervals based on the log hazard.

VALUE:

an object of class coxph. See the coxph.object help file for details.

SIDE EFFECTS:

Survial curves will be drawn if requested. The object save.name will be created or appended to if predictions or standard errors are saved.

SEE ALSO:

, ,