[R] Overlay line on a bar plot - multiple axis

[Jim Lemon]

Hi Miluji,
You have a problem as the x-axis is different for the two plots.
However, the y-axis can be the same, so you really don't need
different axes. I think you want to display city within week. The
example below shows how to display both the counts for the cities by
week, the mean z by week and then use legends to identify the
different parts of the plot. I'm afraid that I didn't even try to do
this in ggplot.

# this uses your structure "dat"
weekcity<-dat[order(dat$week,dat$city),1:4]
barpos<-barplot(matrix(weekcity$count,nrow=5),names.arg=1:3,
 legend.text=c("Akron","Boston","Houston","NYC","OKC"),
 args.legend=list(x=8,y=6),xlab="count",
 ylab="Week",beside=TRUE,horiz=TRUE)
lines(by(weekcity$z,weekcity$week,mean),colMeans(barpos),
 type="b",lwd=2)
legend(5.5,16,"Weekly mean of z",pch=1,lty=1,lwd=2)

Jim

On Mon, Apr 30, 2018 at 4:19 AM, Miluji Sb <milujisb using gmail.com> wrote:
> Dear all,
>
> I am trying to make a similar plot -
> https://peltiertech.com/images/2013-09/BarLineSampleChart4.png.
>
> I have data for two variables; count and z by city and week. I would like
> to have a horizontal bar plot of *count* by city and a line plot of weekly
> average of the variable *z*.
>
> I have tried the following:
>
> ggplot() + geom_bar(data=dat, aes(x=city, y=count),
> stat="identity",position="dodge") + coord_flip() +
> geom_line(data=dat, aes(x=week, y=mean_tmin))
>
> However, this flips the code for both the bar plot and the line plot. I
> would the y-axis of the line plot to be on the left-vertical axis. I read
> that it is difficult to have a secondary axis with ggplot. Any help will be
> highly appreciated. Thank you!
>
> dput(dat)
> structure(list(city = structure(c(4L, 5L, 1L, 3L, 2L, 4L, 3L,
> 1L, 2L, 5L, 3L, 1L, 2L, 4L, 5L), .Label = c("Akron", "Boston",
> "Houston", "NYC", "OKC"), class = "factor"), week = c(1L, 1L,
> 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L), count = c(2.8,
> 2.7, 3.1, 2.5, 3.5, 5.3, 4.5, 9.5, 2.7, 2.1, 4.5, 9.5, 2.7, 5.3,
> 2.1), z = c(-4.1, 1.7, 1.5, 12.8, 1.9, 4, 11.2, 1.4, 2, 4, 10.9,
> 1.4, 1.7, 3.9, 4.3), City = structure(c(4L, 5L, 1L, 3L, 2L, 4L,
> 3L, 1L, 2L, 5L, 3L, 1L, 2L, 4L, 5L), .Label = c("Akron", "Boston",
> "Houston", "NYC", "OKC"), class = "factor"), Count = c(4.466667,
> 2.3, 7.366667, 3.833333, 2.966667, 4.466667, 3.833333, 7.366667,
> 2.966667, 2.3, 3.833333, 7.366667, 2.966667, 4.466667, 2.3),
>     mean_tmin = c(2.76, 2.76, 2.76, 2.76, 2.76, 4.52, 4.52, 4.52,
>     4.52, 4.52, 4.44, 4.44, 4.44, 4.44, 4.44)), .Names = c("city",
> "week", "count", "z", "City", "Count", "mean_tmin"), class = "data.frame",
> row.names = c(NA,
> -15L))
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.