ggplot: Plot for ROC curve

In modelROC: Model Based ROC Analysis

Description

Plot for ROC curve

Usage

## S3 method for class 'auc_coxph'
ggplot(
  data,
  mapping,
  color = NULL,
  lwd = 1.05,
  grid.space = 2,
  ncol = NULL,
  ...,
  environment = parent.frame()
)

## S3 method for class 'roc_coxph'
ggplot(
  data,
  mapping,
  color = NULL,
  lwd = 1.05,
  grid.space = 2,
  rank = FALSE,
  ncol = NULL,
  ...,
  environment = parent.frame()
)

## S3 method for class 'roc_logit'
ggplot(
  data,
  mapping,
  color = NULL,
  lwd = 1.05,
  grid.space = 2,
  rank = FALSE,
  ...,
  environment = parent.frame()
)

Arguments

data	resultes of roc() function
mapping	ignore
color	one or more colors
lwd	logical or integers
grid.space	space between grids, default is 2
ncol	number of column for grid plot
...	ignore
environment	ignore
rank	rank by AUC

Value

a ggplot picture.

Examples

library(ggDCA)
library(rms)
library(modelROC)
###  COX ----------

fit <- cph(Surv(time,status)~ANLN+CENPA+GPR182,LIRI)

####        one model, one time ####
#----            auc for model

r <- auc(fit,
          model='This is model') # one model
unique(r)
ggplot(r)

r <- auc(fit,
          model=TRUE)            # all model
unique(r)
ggplot(r)


#----            auc for x
r <- auc(fit,
          x='ANLN')              # one x
unique(r)
ggplot(r)

r <- auc(fit,
          x=c('ANLN','CENPA'))   # more x
unique(r)
ggplot(r)

r <- auc(fit,
          x=TRUE)                # all x
unique(r)
ggplot(r)

r <- auc(fit,
          model=TRUE,            # one model
          x=TRUE)                # all x
unique(r)
ggplot(r)

r <- auc(fit,
          model='Three Genes',   # specify model names
          x=TRUE)                # all X
unique(r)
ggplot(r)



####        more models ####
fit2 <- cph(Surv(time,status)~ANLN+CENPA,LIRI)


r <- auc(fit,fit2,
          model=c('Three Genes','Two Genes'))            #
unique(r)
ggplot(r)



r <- auc(fit,fit2,
          model=TRUE,
          x=TRUE)
unique(r)
ggplot(r)

library(ggDCA)
library(rms)
fit <- lrm(status~ANLN+CENPA+GPR182,LIRI)
####        one model ####
pp <- roc(fit,
          model=TRUE) # one model
unique(pp)
ggplot(pp)

pp <- roc(fit,
          x='ANLN') # one x
unique(pp)
ggplot(pp)


pp <- roc(fit,
          x=c('ANLN','CENPA')) # more x
unique(pp)
ggplot(pp)


pp <- roc(fit,
          x=TRUE) # ALL x
unique(pp)
ggplot(pp)


pp <- roc(fit,
          model=TRUE, # one model
          x=TRUE) # ALL x
unique(pp)
ggplot(pp)

pp <- roc(fit,
          model='Three Genes', # specify model name
          x=TRUE) # ALL x
unique(pp)
ggplot(pp)

####        more model   ####

fit2 <- lrm(status~ANLN+CENPA,LIRI)
pp <- roc(fit,fit2,
          model=TRUE) # all model
unique(pp)
ggplot(pp)


pp <- roc(fit,fit2,
          model=c('Three Genes','Two Genes')) # specify model names
unique(pp)
ggplot(pp)

pp <- roc(fit,fit2,
          x=TRUE,                             # all x
          model=c('Three Genes','Two Genes')) # all model
unique(pp)
ggplot(pp)

###  COX ----------

fit <- cph(Surv(time,status)~ANLN+CENPA+GPR182,LIRI)
range(LIRI$time)
####        one model, one time ####
#----            roc for model

pp <- roc(fit, times=1,
          model='This is model') # one model
unique(pp)
ggplot(pp)

pp <- roc(fit, times=1,
          model=TRUE)            # all model
unique(pp)
ggplot(pp)

#----            roc for x
pp <- roc(fit, times=1,
          x='ANLN')              # one x
unique(pp)
ggplot(pp)

pp <- roc(fit, times=1,
          x=c('ANLN','CENPA'))   # more x
unique(pp)
ggplot(pp)

pp <- roc(fit, times=1,
          x=TRUE)                # all x
unique(pp)
ggplot(pp)

pp <- roc(fit, times=1,
          model=TRUE,            # one model
          x=TRUE)                # all x
unique(pp)
ggplot(pp)

pp <- roc(fit, times=1,
          model='Three Genes',   # specify model names
          x=TRUE)                # all X
unique(pp)
ggplot(pp)

####        one model, more time ####

pp <- roc(fit, times=c(1,2,3,4,5,6),
          model=TRUE)            # one model
unique(pp)
ggplot(pp)


pp <- roc(fit, times=c(1,2),
          x =  'ANLN')           # one x

unique(pp)
ggplot(pp)


pp <- roc(fit, times=c(1,2,3,4,5,6),
          x = c('ANLN','CENPA')) # more x
unique(pp)
ggplot(pp,ncol = 3)


pp <- roc(fit, times=c(1,2),
          model=TRUE,            # one model
          x = TRUE) # all x
unique(pp)
ggplot(pp)

####        more models, one time ####
fit2 <- cph(Surv(time,status)~ANLN+CENPA,LIRI)

pp <- roc(fit,fit2,times=1,
          x=TRUE,
          model=c('Three Genes','Two Genes'))            #
unique(pp)
ggplot(pp)
####        more models, more time ####
pp <- roc(fit,fit2,times=c(1,2),
          model=c('Three Genes','Two Genes'))            #
unique(pp)
ggplot(pp)

pp <- roc(fit,fit2,times=c(1,2),
          x=TRUE,
          model=c('Three Genes','Two Genes'))            #
unique(pp)
ggplot(pp)

modelROC documentation built on June 25, 2021, 5:09 p.m.