x轴标签乱序

mimisikai

遇到好几次这样的问题，x轴是字符类变量（如这才是日期），xyplot时不是按照原始顺序做图，最后各点之间的线就乱了，之后我将x轴变量套了一个【】，大多数不乱了，但还有两个顺序颠倒了，如下图。

请问大家如何解决？数据已经附上

[attachment=228880,1676]

ming_uld

You'd better give a reproductive example.

nan.xiao

<br />
require(lattice)</p>
<p>x = data.frame(<br />
  time = c(<br />
    '2005/08', '2005/10', '2006/04', '2006/11',<br />
    '2007/04', '2008/03', '2008/12', '2009/06',<br />
    '2009/12', '2010/10', '2011/04'<br />
    ),<br />
  cd4 = c(<br />
    422.9, 380.8, 400.5, 304.0,<br />
    439.9, 538.6, 522.0, 502.9,<br />
    556.9, 504.9, 463.8<br />
    )<br />
)</p>
<p>xyplot(cd4 ~ time, data = x,<br />
       aspect = "xy", type = "o",<br />
       xlab = "Time", ylab = "CD4",<br />
       scales = list(x = list(rot = 45),<br />
                     y = list(alternating = 2)<br />
                     )<br />
       )<br />

[attachment=228893,1678]

YSU

回复第3楼的 nan.xiao：

三楼把2005/8 变成了2005/08，这样2005/8 就比2005/10 小了。

<br />
> class(x$time)<br />
[1] "factor"<br />
> levels(x$time)<br />
 [1] "2005/08" "2005/10" "2006/04" "2006/11" "2007/04" "2008/03" "2008/12"<br />
 [8] "2009/06" "2009/12" "2010/10" "2011/04"<br />
><br />

Excel会把时间自动变成它的格式，比如2005-08-12会显示成8/12/2005，这种变化非常讨嫌，不知道大家是怎么抑制这种转变的。csv文件似乎不能把cell格式设为"text"。

huangruiji110

回复第3楼的 nan.xiao：请问一下版主，有一个csv文件，里面有一列是日期，800多个数据，怎么把这一列以日期的格式显示作为横坐标轴，只取其中一部分日期就可以，全部显示的话很难看

nan.xiao

lattice一般可以自动选择合适数量的刻度标签 ggplot2应该也是可以的

就上例来说要手工指定使用scales中的参数at就可以了配合参数labels还可以显式指定需要显示的刻度文本详见lmdvr pp.137

R基础图形中也有类似参数

<br />
xyplot(cd4 ~ time, data = x,<br />
       aspect = "xy", type = "o",<br />
       xlab = "Time", ylab = "CD4",<br />
       scales = list(x  = list(rot = 45, at = c(1, 2, 8, 11)),<br />
                     y  = list(alternating = 2)<br />
                     )<br />
       )<br />

[attachment=230893,1756]

huangruiji110

回复第6楼的 nan.xiao：

使用xyplot()中出现了两个问题：

1.代码运行出来后纵坐标轴的值不见了，后者说全部集中到了一个小区域

<br />
SZZS_M <- read.csv("SZZS_M.csv", header = TRUE)<br />
xyplot(price ~ date, data = SZZS_M,aspect = "xy", type = "l",<br />
       xlab = "Date", ylab = "Price",<br />
       scales = list(x = list(rot = 45),at = c(1,26,51,76,101,126,151,176,201,226,251)<br />
                     )<br />
       )<br />

2.加了一个abbreviate()指令后，运行出现了错误

<br />
SZZS_M <- read.csv("SZZS_M.csv", header = TRUE)<br />
SZZS_M$date <- abbreviate(SZZS_M$date, 7)<br />
xyplot(price ~ date, data = SZZS_M,aspect = "xy", type = "l",<br />
       xlab = "Date", ylab = "Price",<br />
       scales = list(x = list(rot = 45),at = c(1,26,51,76,101,126,151,176,201,226,251)<br />
                     )<br />
       )<br />

警告信息：

1: In order(as.numeric(x)) : 强制改变过程中产生了NA

2: In diff(as.numeric(x[ord])) : 强制改变过程中产生了NA

3: In function (x, y, type = "p", groups = NULL, pch = if (is.null(groups)) plot.symbol$pch else superpose.symbol$pch, :强制改变过程中产生了NA

麻烦版主帮忙解答一下，先谢谢了

[attachment=231627,1797]

huangruiji110

问了一下老师，稍微改一下就行了

<br />
SZZS <- read.csv("SZZS_M.csv", header = TRUE)<br />
date=abbreviate(SZZS$date,7)<br />
price=SZZS$price<br />
SZZS_M <- data.frame(date,price)<br />
xyplot(price ~ date, data = SZZS_M,aspect = "xy", type = "l",<br />
       xlab = "Date", ylab = "Price",<br />
       scales = list(x = list(rot = 45,at = c(1,26,51,76,101,126,151,176,201,226,251)),<br />
                     y = list(alternating=1)<br />
                     )<br />
       )<br />