请教R语言做矩阵散点图，添加相关系数，并采用稳健回

cor()函数可以提供双变量之间的相关系数，还可以用scatterplotMatrix()函数生成散点图矩阵 不过R语言没有直接给出偏相关的函数； 我们要是做的话，要先调用cortest()对变量进行Pearson相关性分析， 得到简单相关系数，然后做t检验，判断显著性。

用table函数。

在R语言中，table函数用于计算输入变量中各个元素出现的频数。table函数在数据分析中很常用，比如统计样本中各个因素的分类分布情况，或者计算某个离散变量的各取值位置分别出现了多少次。

table函数的基本语法如下：table（x），其中，x是一个向量或因子，在这个向量或因子中我们需要计算每个元素出现的频率。

数据准备

统计概念学习。

(1)当观测值为奇数时，(n+1)/2位置的观测值即为中位数。

(2)当观测值个数为偶数时，n/2和n/2 + 1位置的两个观测值的平均数为中位数。

数据的标准化

数据的标准化是指将数据按照比例缩放，使之落入一个特定的区间。

数据的中心化：数据集中的各项数据减去数据集的均值。 中心化公式：

数据的标准化：中心化之后的数据再除以数据集的标准差，即数据集中的各项数据减去数据集的均值再除以数据集的标准差。经过该方法处理的数据的均值是0，标准差是1。 标准化公式：

其中，x为观测者， 为平均值，S为标准差。

scale(x, center = TRUE, scale = TRUE)

x：用于标准化的数据，

center=T：表示数据中心化，

scale=T：表示数据标准化。

默认情况下，center=TRUE，scale=TRUE，scale()函数首先把一组数的每个数都减去这组数的平均值，然后除以这组数的均方根。

如果scale=TRUE，而center=FALSE，那么，scale()函数不会把一组数中的每个数减去平均值，而直接除以这组数据的均方根。

参考资料：

可以使用数据标号“text()”函数text()函数跟在画图函数语句后面，即先画出图，再标号。

下面为来自R的text()函数使用方法（疑难词汇已经标出）

Description

text draws the strings given in the vector(矢量) labels at the coordinates(坐标) given by x and y y may be missing since xycoords(x, y) is used for construction of the coordinates

Usage

text(x, )

## Default S3 method:

text(x, y = NULL, labels = seq_along(x$x), adj = NULL,pos = NULL, offset = 05, vfont = NULL,cex = 1, col = NULL, font = NULL, )

Arguments

x, y

numeric(数) vectors(矢量) of coordinates(坐标) where the text labels should be written If the length of x and y differs, the shorter one is recycled

labels

a character vector or expression specifying the text to be written An attempt is made to coerce(强制) other language objects (names and calls) to expressions, and vectors and other classed objects to character vectors byascharacter If labels is longer than x and y, the coordinates(坐标) are recycled to the length of labels

adj

one or two values in [0, 1] which specify(指定) the x (and optionally(可选择的) y) adjustment(调整) of the labels(标签) On most devices(装置) values outside that interval will also work

pos

a position specifier for the text If specified this overrides(代理佣金) any adj value given Values of 1, 2, 3 and 4, respectively(分别地) indicate(表明) positions below, to the left of, above and to the right of the specified coordinates

offset

when pos is specified(指定), this value gives the offset(抵消) of the label(标签) from the specified coordinate(坐标) in fractions(分数) of a character width

vfont

NULL for the current font family, or a character vector(矢量) of length 2 for Hershey vector fonts The first element(元素) of the vector selects a typeface and the second element selects a style Ignored(驳回诉讼) if labels is an expression

cex

numeric character expansion factor(因素); multiplied by par("cex") yields(产量) the final character size NULL and NA are equivalent to 10

col, font

the color and (if vfont = NULL) font to be used, possibly vectors(矢量) These default to the values of the global graphical parameters in par()

further graphical parameters (from par), such as srt, family and xpd

Details

labels must be of type character or expression (or be coercible(可强迫的) to such a type) In the latter case, quite a bit of mathematical(数学的) notation(符号) is available such as sub- and superscripts(上标), greek letters,fractions(分数), etc

adj allows adjustment of the text with respect to (x, y) Values of 0, 05, and 1 specify(指定) left/bottom, middle and right/top alignment(队列), respectively(分别地) The default is for centered text, ie, adj = c(05, NA)Accurate(精确的) vertical(垂直的) centering needs character metric(度量标准) information on individual(个人的) characters which is only available on some devices(装置) Vertical alignment is done slightly differently for character strings and for expressions: adj = c(0,0) means to left-justify and to align(结盟) on the baseline for strings but on the bottom of the bounding box for expressions This also affects vertical(垂直的) centering: for strings the centeringexcludes(排除) any descenders(下降) whereas(然而) for expressions it includes them Using NA for strings centers them, including descenders

The pos and offset arguments can be used in conjunction(结合) with values returned by identify to recreate(再创造) an interactively(交互式地) labelled(贴上标签的) plot(情节)

Text can be rotated(旋转的) by using graphical parameters srt (see par); this rotates about the centre set by adj

Graphical parameters col, cex and font can be vectors(矢量) and will then be applied cyclically(周期的) to the labels (and extra values will be ignored(驳回诉讼)) NA values of font are replaced by par("font"), and similarly for col

Labels whose x, y or labels value is NA are omitted(省略) from the plot(情节)

What happens when font = 5 (the symbol(象征) font) is selected can be both device- and locale-dependent Most often labels will be interpreted(说明) in the Adobe symbol encoding, so eg "d" is delta, and "\300" is aleph

Euro symbol

The Euro symbol may not be available in older fonts In current versions of Adobe symbol fonts it is character 160, so text(x, y, "\xA0", font = 5) may work People using Western European locales(场所) on Unix-alikes can probably select ISO-8895-15 (Latin-9) which has the Euro as character 165: this can also be used for postscript and pdf It is \u20ac in Unicode, which can be used in UTF-8 locales(场所)

In all the European Windows encodings the Euro is symbol(象征) 128 and \u20ac will work in all locales: however not all fonts will include it It is not in the symbol font used for windows and related devices(装置), including the Windows printer

References

Becker, R A, Chambers, J M and Wilks, A R (1988) The New S Language Wadsworth & Brooks/Cole

Murrell, P (2005) R Graphics Chapman(叫卖小贩) & Hall/CRC Press

See Also

textformula for the formula(公式) method; mtext, title, Hershey for details on Hershey vector(矢量) fonts, plotmath for details and more examples on mathematical(数学的) annotation(注释)

Examples

plot(-1:1, -1:1, type = "n", xlab = "Re", ylab = "Im")

K <- 16; text(exp(1i 2 pi (1:K) / K), col = 2)

## The following two examples use latin1 characters: these may not

## appear correctly (or be omitted entirely)

plot(1:10, 1:10, main = "text() examples\n~~~~~~~~~~~~~~",

sub = "R is GNU ©, but not ® ")

mtext("«Latin-1 accented chars»: éè øØ å<Å æ<Æ", side = 3)

points(c(6,2), c(2,1), pch = 3, cex = 4, col = "red")

text(6, 2, "the text is CENTERED around (x,y) = (6,2) by default",

cex = 8)

text(2, 1, "or Left/Bottom - JUSTIFIED at (2,1) by 'adj = c(0,0)'",

adj = c(0,0))

text(4, 9, expression(hat(beta) == (X^t X)^{-1} X^t y))

text(4, 84, "expression(hat(beta) == (X^t X)^{-1} X^t y)",

cex = 75)

text(4, 7, expression(bar(x) == sum(frac(x[i], n), i==1, n)))

## Two more latin1 examples

text(5, 102,

"Le français, c'est façile: Règles, Liberté, Egalité, Fraternité")

text(5, 98,

"Jetz no chli züritüütsch: (noch ein bißchen Zürcher deutsch)")