R和ggplot-图分布和线

Question

I have a question that seems very simple, but I cannot seem to figure out. I have a dataset with treatments in a given year. There are 3 different treatments. I would like to create two plots:

One that looks like this:

And one that looks like this:

, except, I would like to stack multiple treatments (three instead of just the one in the example).

Let's say we have the follow df:

y=c(2001,2001,2001,2001,2002,2002,2002,2003,2003,2003,2003,2004,2004)
t=c("a","a","b","c","a","a","b","c","a","a","b","c","b")
df=data.frame(y,t)

I've tried using

geom_plot()

But it does not work. The closest I could get to have R do the proportions for me is the following stacked histogram using code from another post:

p+geom_histogram(aes(y=..density.., color=t , fill=t))

Answer 1

For the types of charts you show, you'll need to compute the proportions before you plot. The table function can be used to do the counts of t by year and t . ave with sum by y then computes the annual sums for the proportions. Your first plot is made with geom_area while the second is a standard line and point plot. The code could look like

library(ggplot2)
y=c(2001,2001,2001,2001,2002,2002,2002,2003,2003,2003,2003,2004,2004)
t=c("a","a","b","c","a","a","b","c","a","a","b","c","b")
df=data.frame(y, t)

# Count number of t's by year 
  df_tab <- as.data.frame(table(df), stringsAsFactors=FALSE)
# convert counts to percents
  df <-  data.frame(df_tab, p=df_tab$Freq/ave(df_tab$Freq, df_tab$y, FUN=sum))
  df$y <- as.numeric(df$y)
# Set plot colors and themes
  plot_colours <- c(a="red3", b = "orange", c = "blue")
  plot_theme <- theme(axis.title = element_text(size = 18 )) +
                 theme(axis.text = element_text(size = 18)) +
                 theme(legend.position="top", legend.text=element_text(size=18))
# make area plot
  sp <- ggplot(data=df, aes(x=y, y= 100*p, fill=t)) + geom_area()
  sp <- sp + scale_fill_manual(values=plot_colours)
  sp <- sp + labs(x="Year", y = "Percentage of Patients")
  sp <- sp + plot_theme
  plot(sp)

# make line plot
  sp <- ggplot(data=df, aes(x=y, y=p, colour=t))
  sp <- sp + geom_line(aes(ymax=1), position="stack", size=1.05) + geom_point(aes(ymax=1), position="stack", size=4)
  sp <- sp + scale_colour_manual(values=plot_colours)
  sp <- sp + labs(x="Year", y = "Proportion Receiving Treatment")
  sp <- sp + plot_theme
  plot(sp)

which produces the plots

and