Multiple ROC curves in one graph

[alexgalli]

Dear R Studio Community,

I am trying to plot 2 ROC curves in one graph to nicely compare them.

I used the "cutpointr" package and I don't know how to merge the 2 results.

 y <- data[ ,  c( " pnf " ,  " lac " )]
 
 roc1 <- cutpointr( data = y , x= lac , class = pnf , 
                 method = maximize_metric, metric = sum_sens_spec, na.rm = FALSE)

 summary(roc1)
 plot(roc1)



 y <- data[ ,  c( " aed " ,  " lac " )]
 
 roc2 <- cutpointr( data = y , x= lac , class = aed , 
                 method = maximize_metric, metric = sum_sens_spec, na.rm = FALSE)

 summary(roc2)
 plot(roc2)

I read some previous post suggestiong to use pROC package too, and so I did

pred1 <- ifelse ( data$aed == TRUE, 1, 0)
 
 r.lac <- roc(pred1, data$lac)

but I kept stopping with an ERROR : "Error: Only strings can be converted to symbols".

Can you give me any suggestion on how to move forward?

Thank you!

[jrkrideau]

We need a reproducible example (reprex) and especially some sample data.

FAQ: How to do a minimal reproducible example ( reprex ) for beginners

A minimal reproducible example consists of the following items: A minimal dataset, necessary to reproduce the issue The minimal runnable code necessary to reproduce the issue, which can be run on the given dataset, and including the necessary information on the used packages. Let's quickly go over each one of these with examples: Minimal Dataset (Sample Data) You need to provide a data frame that is small enough to be (reasonably) pasted on a post, but big enough to reproduce your issue. Let's say, as an example, that you are working with the iris data frame head(iris) #> Sepal.Length Sepal.Width Petal.Length Petal.Width Species #> 1 5.1 3.5 1.4 0.…

A handy way to supply sample data is to use the dput() function. See ?dput. If you have a very large data set then something like head(dput(myfile), 100) will likely supply enough data for us to work with.

[arthur.t]

library(pROC)
#> Type 'citation("pROC")' for a citation.
#> 
#> Attaching package: 'pROC'
#> The following objects are masked from 'package:stats':
#> 
#>     cov, smooth, var
library(tidyverse)

# based on example from roc documentation
# data(aSAH)
# roc(aSAH$outcome, aSAH$s100b,
#     levels=c("Good", "Poor"))
# roc(aSAH$outcome, aSAH$s100b,
#     levels=c("Good", "Poor")) %>% plot()

# make a data set  
data(aSAH)
my_data <-aSAH %>% mutate(pred1 = s100b, 
                          # add a second prediction
                          pred2 = pmax(0, pred1 + runif(length(pred1), 0, .2)))

extract_roc_df <- function(roc_obj) {
  # extract sens and spec from rob_object as a tibble
  as_tibble(roc_obj[c("sensitivities", "specificities")])
}

# get roc curves for each predictor
map2(my_data[, c("pred1", "pred2")],
     list(my_data$outcome),
     ~roc(.y, .x, levels=c("Good", "Poor"))) %>%
  # extract sens and spec as data frames and bind rows
  map_dfr(extract_roc_df, .id = "pred") %>%
  # plot
  ggplot() + 
  geom_abline(slope = 1, intercept = 1, color = "darkgrey") +
  aes(specificities, sensitivities, color = pred) + 
  geom_line() + 
  scale_x_reverse()
#> Setting direction: controls < cases
#> Setting direction: controls < cases

^{Created on 2021-08-12 by the reprex package (v1.0.0)}