R：固定x轴缩放并转换plot（）xy参数到points（）

Question

我有一个data.frame xy，我正如下面的代码所示。

我是否有办法将xy plot（）中的xx和yy转换为points()命令，以便我可以设置type='n'并添加{在段命令后{1}}为了更好地控制它？

points()

如果可能的话，如果x轴可以处于固定范围（例如从1940到2014），并且如果存在1940之前的值，那么x轴应该是自动的。 y轴的范围总是不同的。我怎么能在我的代码中加入它？

Answer 1

xy <- data.frame(NAME = c("NAME1", "NAME1", "NAME1", "NAME2", "NAME2", "NAME2"),
                 ID = c(87, 87, 87, 199, 199, 199), 
                 X_START_YEAR = c(1984, 1986, 1984, 1899, 1909, 1924),
                 Y_START_VALUE = c(75, 25, -90, -8, -55, -10),
                 X_END_YEAR = c(1986, 1994, 1999, 1909, 1924, 1927), 
                 Y_END_VALUE = c(20, 50, -15, -70, -80, -100))
xy

ind <- split(xy, xy$ID)

for (i in ind){

  xx = unlist(i[, grep('X_', colnames(i))])

  yy = unlist(i[, grep('Y_', colnames(i))])    

  fname <- paste0(i[1, 'ID'], '.png')

  png(fname, width = 1679, height = 1165, res = 150)

  # test for xx smaller than 1940
  if(any(xx < 1940)) {

    my_x_lim <- c(min(xx), max(xx))

  } else {

    my_x_lim <- c(1940, 2014)}

  # plot the data using pch at your choice and color them as you like
  # plot your x limits
  par(mar = c(6, 8, 6, 5))

  plot(xx, 
       yy,
       main = unique(i[, 1]),
       xlab = "Time [Years]",
       ylab = "Value [m]",
       pch = 21, col = "black",
       xlim = my_x_lim)

  axis(2, at = seq(-100000, 100000, 500), cex.axis = 1, labels = FALSE, tcl = -0.3)

  i <- i[, -1]

  segments(i[, 2], i[, 3], i[, 4], i[, 5],lwd = 2)

  points(xx, yy, pch = 21, col = "black")

  dev.off()
}

Answer 2

ggplot2库有一个分段工具，可以将这些数据绘制得非常紧凑，因此易于维护和调整以供将来参考。在这里。

它基本上是七行，但可以进一步压缩。请注意修改x轴的额外灵感。

dat = xy

# Adapted from the other solution
if(any(dat$X_START_YEAR < 1940)) {  
  my_x_lim <- c(min(dat$X_START_YEAR), max(dat$X_END_YEAR))
  } else {
    my_x_lim <- c(1940, 2014)
  }

# plot based on Hadley Wickham's ggplot2
library(ggplot2)
p = ggplot(dat)
p = p + geom_segment(aes(x = X_START_YEAR, xend = X_END_YEAR, y = Y_START_VALUE, yend = Y_END_VALUE))
# p = p + facet_wrap(~ID)
# proposal to uncomment if y-axis should not be equivalent over multiple facets (makes the plot harder to read, though):
p = p + facet_wrap(~ID, scales = "free")
# Alternatively, only let the x-axis scale vary freely (this gets you ride of the initial procedure to determine the x-axis limits)
# p = p + facet_wrap(~ID, scales = "free_x")
p = p + xlab("Time [Years]") + ylab("Value [m]")
p = p + xlim(my_x_lim)
print(p)