How can I plot the cumulative probability of an event occuring from a dataframe over time?

Question

I want to find a general way to find the cumulative probability of an event using a dataframe output. For example, say I have a model for which I have run 3 simulations from days 0-5 which outputs this dataframe "hospital":

Node	Time	Hospitalised
1	0	0
2	0	0
3	0	0
1	1	0
2	1	1
3	1	0
1	2	0
2	2	0
3	2	1
1	3	0
2	3	1
3	3	3
1	4	1
2	4	1
3	4	0
1	5	0
2	5	0
3	5	0

I want to find and plot the cumulative probability that at least 1 person has been hospitalised over time. It is cumulative in that for each time point I care about whether there has ever been anyone in hospital for that particular simulation (currently or before the current time). The probability is for >0 hospitalised / total # simulations.

This would be the output for this simplified example

Answer 1

Here's a tidyverse solution:

library(tidyverse)

data %>% 
  group_by(Node) %>%
  mutate(any_hospitalised = sign(cumsum(Hospitalised))) %>%
  group_by(Time) %>%
  summarize(probability = mean(any_hospitalised)) %>%
  ggplot(aes(Time, probability)) +
  geom_line() +
  geom_point() +
  theme_bw()

Although you may prefer a step plot to a line plot in this scenario:

data %>% 
  group_by(Node) %>%
  mutate(any_hospitalised = sign(cumsum(Hospitalised))) %>%
  group_by(Time) %>%
  summarize(probability = mean(any_hospitalised)) %>%
  ggplot(aes(Time, probability)) +
  geom_step() +
  geom_point() +
  theme_bw()

^{Created on 2022-03-06 by the reprex package (v2.0.1)}