getDCA2: ggplot style for decision estimate analysis

In huangwb8/luckyExperiment: useful functions for biology experiment

ggplot style for decision estimate analysis

Description

ggplot style for decision estimate analysis

Usage

getDCA2(
  data,
  group.by = NULL,
  time.col = "time",
  status.col = "event",
  clusters,
  time.raw.type = c("Day", "Month", "Year")[1],
  time.target.type = c("Day", "Month", "Year")[3],
  time.knot = c(3, 5),
  study.design = "cohort",
  population.prevalence = c(NA, NA),
  load.existed.res = F,
  plot.ntbft.by = c("group", "model")[1],
  plot.ntbft.ymin = NULL,
  plot.ntbft.ymax = NULL,
  width = 9,
  height = 7,
  add.bootc = T,
  add.cv = F,
  add.ci = T,
  plot.legend.position = "right",
  plot.type = c("time", "cluster", "disperse")[1],
  plot.label,
  parameter.label = c(all = "All positive", none = "All negative"),
  plot.family = "Arial",
  names = "project"
)

Arguments

data	a data frame with survival data(time & event)
group.by	group colname
time.col	colname of time value
status.col	Numeric.status.col must be a binary parameter with 0(event not happened yet) or 1(event happened) value.
clusters	list of characters.Some of the colnames of the data representing the value involving the glm model.
time.raw.type	one of "Day","Month","Year"
time.target.type	one of "Day","Month","Year"
time.knot	time knot to create survival probability
study.design	Either 'cohort' (default) or 'case-control' describing the study design used to obtain data. See details for more information.
population.prevalence	Outcome prevalence rate in the population used to calculate decision curves when study.design = 'case-control'.
load.existed.res	whether to use existed rds result in the work space
plot.ntbft.ymin	the ymin of plot.ntbft. Default is ymin=-0.05
plot.ntbft.ymax	the ymax of plot.ntbft. Default is ymax=max(nb[,c(nolines,nobands)],na.rm=T)
plot.legend.position	the position of plot legend
plot.type	Character. One/some of c("time","cluster","disperse"). if plot.type="time",then multiple clusters would be ploted for every time; if plot.type="cluster", then multiple times would be ploted for every cluster; if plot.type="disperse", then every single plot would be ploted indivitually.
plot.label	List. The name of list are the paste0 of elements in one of clusters with "-".The content of the list are the real names that would be printed in the plot legend/title.
parameter.label	the label of parameter label.
names	part of save files or plots.

Details

beta version

Value

ggplot style plot and result

Author(s)

Weibin Huang<[e-mail]>

Examples

## This is a simulative process and available only with CORRECT VARIABLES

huangwb8/luckyExperiment documentation built on June 29, 2023, 10:15 a.m.