Qfyilyi

I want to create a set of histograms that will make use of facet_wrap. The y limits vary a lot, sometimes there is no data for a given combination. So, I thought I could control the y limits using a dummy data frame and geom_blank.

Geom_blank() works as expected when I use geom_point() but not when when I use geom_histogram()

For example, I create these two data frames

test  <- data.frame(grp=rep(c("a", "b"),3), x.value=rep(c(10,20,30), each=2), 

         y.value=rep(1:3, each=2))

dummy <- data.frame(grp=rep(c("a", "b"),3),  x.value=rep(c(10,20,30), each=2),  

         y.value=c(0,10,10,20,20,30))  

If I want to do a scatterplot:

p1 <- ggplot(test, aes(x=x.value, y=y.value)) + geom_point()

p1 <- p1 + facet_wrap(~grp, scales="free")

p1



p2 <- p1 + geom_blank(data=dummy)

p2  

geom_blank() changed the y limits.

But when I try to create a histogram

p3 <- ggplot(test, aes(x=x.value, weight=y.value)) + geom_histogram(bins=6)

p3 <- p3 + facet_wrap(~grp, scales="free")

p3



p4 <- p3 + geom_blank(data=dummy)

p4

geom_blank() has no impact.

How can I control complex y limits on a set of histograms?

You need to map y.value to y like you did in p1 and p2.

p4 <- p3 + geom_blank(aes(y = y.value), data=dummy)

p4