怎样生成这样的时间序列图？

我有数据，见附件：
x1.zip (459 Bytes) 本附件包括：

• x1.csv

2012-3-14 10:45:26 上传

想用R生成下面这样的时间序列图（不是一个图一个图单作，而是像下图一样包含在一张图中！）：

请问该如何做？

耕耘使者 发表于 2012-3-14 19:54
太好了，但能否让y1等离纵轴近一些？或者让他们大一些？
因为看不清楚是哪个模型的了

rm(list=ls())
par(mfrow=c(3,2),mar=c(0,4,0,2), oma = c(3, 0, 2, 0))
data=read.csv("x1.csv",header=T)
attach(data)
plot(year,y1,xaxt="n", xlab = "", type = "l")
plot(year,y2,xaxt="n", xlab = "", type = "l")
plot(year,y3,xaxt="n", xlab = "", type = "l")
plot(year,y4,xaxt="n", xlab = "", type = "l")
plot(year,y5, xlab = "", type = "l")
plot(year,y7, xlab = "", type = "l")


另外距离近一点的话还是问版主吧，呵呵

看不大清楚图的坐标值。给你个参考，可以对应着y1~y7做你想要的图。
我习惯读txt格式数据，所以把csv转化成txt.
R code:
da=read.table("x1.txt",header=T)
plot(da$year,da$y1,type="l",ylab="agriculture",xlab="time")

universe110 发表于 2012-3-14 11:07
看不大清楚图的坐标值。给你个参考，可以对应着y1~y7做你想要的图。
我习惯读txt格式数据，所以把csv转化成 ...

不是这样的，是如附图那样，同时做所有的变量图，最后显示在一张图上

x1 <- rnorm(30)
x2 <- rnorm(30)
x3 <- rnorm(30)
x4 <- rnorm(30)
x5 <- rnorm(30)

x <- cbind(c(1,2,3), c(4,5,6))
layout(x, c(1,1), c(3,2,3))
layout.show(5)

par(mar=c(0,5,5,2), oma = c(0, 0, 3, 0))
plot(x1, xaxt = "n", xlab = "", type = "l")
par(mar=c(0,5,0,2))
plot(x2, xaxt = "n", xlab = "", type = "l")
par(mar=c(5,5,0,2))
plot(x3, xlab = "Time", type = "l")
par(mar=c(0,5,5,2))
plot(x4, xaxt = "n", xlab = "", type = "l")
par(mar=c(0,5,0,2))
plot(x5, xlab = "Time", type = "l")
mtext("Time", side = 1, cex = 0.75, line = 3)
mtext("Big Title", outer = TRUE, cex = 1)

qoiqpwqr 发表于 2012-3-14 11:28
x1

版主，我现在急着要用，一行一行学已经来不急了，你能否按附件中的数据文件把代码写出来？是六个变量。
呵呵，我太笨了

dat <- read.csv("x1.csv", header = T, sep = ",")
attach(dat)
x <- cbind(c(1,2,3), c(4,5,6))
layout(x, c(1,1), c(4,3,4))

par(mar=c(0,5,4,2), oma = c(0, 0, 3, 0))
plot(year, y1, xaxt = "n", xlab = "", type = "l")
par(mar=c(0,5,0,2))
plot(year, y2, xaxt = "n", xlab = "", type = "l")
par(mar=c(5,5,0,2))
plot(year, y3, xlab = "Time", type = "l")
par(mar=c(0,5,4,2))
plot(year, y4, xaxt = "n", xlab = "", type = "l")
par(mar=c(0,5,0,2))
plot(year, y5, xaxt = "n", xlab = "", type = "l")
par(mar=c(5,5,0,2))
plot(year, y7, xlab = "Time", type = "l")
mtext("Big Title", outer = TRUE, cex = 1)

qoiqpwqr 发表于 2012-3-14 11:49
dat

直接运行的错误是：
> plot(year, y1, xaxt = "n", xlab = "", type = "l")
错误于as.POSIXlt.numeric(x) : 'origin' must be supplied
> par(mar=c(0,5,0,2))
> plot(year, y2, xaxt = "n", xlab = "", type = "l")
错误于as.POSIXlt.numeric(x) : 'origin' must be supplied
> par(mar=c(5,5,0,2))
> plot(year, y3, xlab = "Time", type = "l")
错误于as.POSIXlt.numeric(x) : 'origin' must be supplied
> par(mar=c(0,5,4,2))
> plot(year, y4, xaxt = "n", xlab = "", type = "l")
错误于as.POSIXlt.numeric(x) : 'origin' must be supplied
> par(mar=c(0,5,0,2))
> plot(year, y5, xaxt = "n", xlab = "", type = "l")
错误于as.POSIXlt.numeric(x) : 'origin' must be supplied
> par(mar=c(5,5,0,2))
> plot(year, y7, xlab = "Time", type = "l")
错误于as.POSIXlt.numeric(x) : 'origin' must be supplied
> mtext("Big Title", outer = TRUE, cex = 1)
错误于mtext("Big Title", outer = TRUE, cex = 1) : 还没有调用plot.new

修改程序：用dat$year代替了year,
新的提示是：
> par(mar=c(0,5,4,2), oma = c(0, 0, 3, 0))
> plot(dat$year, y1, xaxt = "n", xlab = "", type = "l")
错误于xy.coords(x, y, xlabel, ylabel, log) :
  'x' and 'y' lengths differ
> par(mar=c(0,5,0,2))
> plot(dat$year, y2, xaxt = "n", xlab = "", type = "l")
错误于xy.coords(x, y, xlabel, ylabel, log) :
  'x' and 'y' lengths differ
> par(mar=c(5,5,0,2))
> plot(dat$year, y3, xlab = "Time", type = "l")
错误于xy.coords(x, y, xlabel, ylabel, log) :
  'x' and 'y' lengths differ
> par(mar=c(0,5,4,2))
> plot(dat$year, y4, xaxt = "n", xlab = "", type = "l")
错误于xy.coords(x, y, xlabel, ylabel, log) :
  'x' and 'y' lengths differ
> par(mar=c(0,5,0,2))
> plot(dat$year, y5, xaxt = "n", xlab = "", type = "l")
错误于xy.coords(x, y, xlabel, ylabel, log) :
  'x' and 'y' lengths differ
> par(mar=c(5,5,0,2))
> plot(dat$year, y7, xlab = "Time", type = "l")
错误于xy.coords(x, y, xlabel, ylabel, log) :
  'x' and 'y' lengths differ
> mtext("Big Title", outer = TRUE, cex = 1)
错误于mtext("Big Title", outer = TRUE, cex = 1) : 还没有调用plot.new
>

麻烦版主了！