Adding v-lines to plot in base R using time series object-CodePudding

I am trying to add vertical lines to a time series plot I made in base R plot(data1,type = 'l',lwd = 1.5, family = "A", ylab ="", xlab = "", main = ""). This plot has a total of 5 plots inside of it. There are two x-axes that are the same (see current plot)

When adding vlines with abline(v=c(27,87, 167, 220, 280, 329), lty=2) I get this result

Is there a way to get them to go on the graphs so it looks something like this but with dashed lines and the lines within the plots.
Or if you know of a better way to plot this in ggplot that would be fantastic as well. Thank you so much in advance.

CodePudding user response：

Here is a toy example of using ggplot to put in vlines.

library(tidyverse)


iris2 <- iris %>% pivot_longer(cols=Sepal.Length:Petal.Length)


ggplot(iris2, aes(x = Petal.Width, y = value))  
  geom_line()  
  facet_wrap(~name, scales="free_y", ncol=2)  
  geom_vline(xintercept=c(.25, .75, 1.25, 1.75, 2.25), 
             linetype='dashed', col = 'blue')

CodePudding user response：

plot calls plot.ts. This has a panel= argument in which you may define a function what should happen in the panels. Since you want a grid and the usual lines, you can do quite easily:

panel <- function(...) {grid(col=1, ny=NA, lty=1); lines(...)}
plot(z, panel=panel)

grid usually uses the axTicks and you can define number of cells, see ?grid.

This also works with abline.

panel <- function(...) {abline(v=seq(1961, 1969, 2), col=1,lty=1); lines(...)}
plot(z, panel=panel)

Data:

set.seed(42)
z <- ts(matrix(rnorm(300), 100, 5), start = c(1961, 1), frequency = 12)