Question

Really struggling with this.

a & b are actual datasets in the real world, a being extremely large. I get an error ggplot2 doesn’t know how to deal with data of class uneval. What I’m trying to do is overlay a single point from a second data set on to the boxplot to highlight how one particular sample compared to a universe.

Any idea what I’m doing wrong? How can I fix it?

a = data.frame(YTD.Retn=runif(1000,-10,10),sector="a")
a = rbind(a,data.frame(YTD.Retn=runif(1000,-10,10),sector="b"))
a = rbind(a,data.frame(YTD.Retn=runif(1000,-10,10),sector="c"))
a = rbind(a,data.frame(YTD.Retn=runif(1000,-10,10),sector="d"))
a = rbind(a,data.frame(YTD.Retn=runif(1000,-10,10),sector="e"))
a = rbind(a,data.frame(YTD.Retn=runif(1000,-10,10),sector="f"))
a = rbind(a,data.frame(YTD.Retn=runif(1000,-10,10),sector="g"))

b = data.frame(sector=c("a","b","c","d","e","f","g"),YTD.Retn=c(5,6,7,3,2,-1,-5))

p1 =ggplot(a,aes(factor(sector),YTD.Retn,fill=factor(sector))) + geom_boxplot() +
  scale_fill_discrete(guide=F) +
  geom_point(b,aes(factor(sector),YTD.Retn))

plot(p1)

Answer 1

You need to name the argument data when called within a geom_...() call. Naming arguments is good practice in general (if somewhat timeconsuming)

p1 =ggplot(data = a, aes(x = factor(sector), y = YTD.Retn, fill=factor(sector))) + 
   geom_boxplot() +
  scale_fill_discrete(guide=F) +
  geom_point(data = b, aes(x= factor(sector),y= YTD.Retn))

plot(p1)