icpack (version 0.1.0)

plot.predict.icfit: Plot method for an object of class `predict.icfit`

Description

Plot method for an object of class `predict.icfit`

Usage

# S3 method for predict.icfit
plot(
  x,
  type = c("hazard", "cumhazard", "survival", "probability"),
  conf.int = TRUE,
  fill = TRUE,
  fillcol = "lightgrey",
  ylim = NULL,
  title = NULL,
  xlab = NULL,
  ylab = NULL,
  selection = NULL,
  nrow = NULL,
  ncol = NULL,
  do_plot = TRUE,
  ...
)

Value

A ggplot grob, containing the plot. Use print() or plot() to show it Multiple objects can be combined by using functions in the package gridExtra.

Arguments

x

The object of class 'predict.icfit' to be plotted

type

Type of plot. Accepted choices: 'hazard' (default), 'cumhazard', 'survival' or 'probability'

conf.int

If `TRUE` a 100*(1 - alpha) percent confidence interval is plotted

fill

Fill area between lower and upper

fillcol

The color for filling (default 'lightgrey')

ylim

The y-limits for the plot

title

Optional title string, or, if x is a list, obtained from 'predict.icfit' using `newdata`, a vector of title strings

xlab

Text for x-label

ylab

Text for y-label

selection

If x is a list, obtained from 'predict.icfit' using `newdata`, then a vector containing the subset of list elements to be plotted, default is to plot all elements of the list

nrow

If x is a list, obtained from 'predict.icfit' using `newdata`, then a number specifying the number of rows to plot; default the square root of the number of list elements to be plotted

ncol

If x is a list, obtained from 'predict.icfit' using `newdata`, then a number specifying the number of columns to plot; default the square root of the number of list elements to be plotted

do_plot

Boolean indicating whether or not to actually plot (default is TRUE)

...

other graphical parameters to be passed on

Examples

# \donttest{
icf <- icfit(Surv(left, right, type='interval2') ~ period + gender + age, data=drugusers)
pred_icf <- predict(icf)
plot(pred_icf)
library(ggplot2)
plot(icf) + xlim(0, 200) + ylim(0, 0.05)
ndata <- drugusers[1:4, ]
pred_nd_icf <- predict(icf, newdata=ndata)
plot(pred_nd_icf) # plot all four
plot(pred_nd_icf[[2]]) # plot only the second
plot(pred_nd_icf, type = "cumhazard") # plot four cumulative hazard curves
plot(pred_nd_icf[[3]], type = "prob", ylim = c(0, 1)) # plot probability curve for nr 3
plot(pred_nd_icf[[4]], type = "surv", ylim = c(0, 1)) # plot survival curve for nr 4
# }