当前位置：文江博客话题详情

r ggplot2 plot correlation

如何将相关矩阵绘制成图表？

发布于 2024-10-27 01:29:43 字数 136 浏览 4 评论 0 原文

我有一个带有一些相关值的矩阵。现在我想在一个看起来或多或少像这样的图表中绘制它：

在此处输入图像描述

我怎样才能实现这一点？

原文

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

幼儿园老大 2024-11-03 01:29:43

看起来“更少”，但值得检查（因为提供了更多视觉信息）：

相关矩阵椭圆：

相关矩阵圆：

请在 corrplot vignette 由下面的 @assylias 引用。

回复收藏 0 原文

痴情换悲伤 2024-11-03 01:29:43

快速、肮脏、大致目标：

library(lattice)

#Build the horizontal and vertical axis information
hor <- c("214", "215", "216", "224", "211", "212", "213", "223", "226", "225")
ver <- paste("DM1-", hor, sep="")

#Build the fake correlation matrix
nrowcol <- length(ver)
cor <- matrix(runif(nrowcol*nrowcol, min=0.4), nrow=nrowcol, ncol=nrowcol, dimnames = list(hor, ver))
for (i in 1:nrowcol) cor[i,i] = 1

#Build the plot
rgb.palette <- colorRampPalette(c("blue", "yellow"), space = "rgb")
levelplot(cor, main="stage 12-14 array correlation matrix", xlab="", ylab="", col.regions=rgb.palette(120), cuts=100, at=seq(0,1,0.01))

在此处输入图像描述

Quick, dirty, and in the ballpark:

library(lattice)

#Build the horizontal and vertical axis information
hor <- c("214", "215", "216", "224", "211", "212", "213", "223", "226", "225")
ver <- paste("DM1-", hor, sep="")

#Build the fake correlation matrix
nrowcol <- length(ver)
cor <- matrix(runif(nrowcol*nrowcol, min=0.4), nrow=nrowcol, ncol=nrowcol, dimnames = list(hor, ver))
for (i in 1:nrowcol) cor[i,i] = 1

#Build the plot
rgb.palette <- colorRampPalette(c("blue", "yellow"), space = "rgb")
levelplot(cor, main="stage 12-14 array correlation matrix", xlab="", ylab="", col.regions=rgb.palette(120), cuts=100, at=seq(0,1,0.01))

enter image description here

回复收藏 0 原文

赠佳期 2024-11-03 01:29:43

使用lattice::levelplot非常简单：

z <- cor(mtcars)
require(lattice)
levelplot(z)

在此处输入图像描述

Very easy with lattice::levelplot:

z <- cor(mtcars)
require(lattice)
levelplot(z)

enter image description here

回复收藏 0 原文

二手情话 2024-11-03 01:29:43

ggplot2 库可以使用 geom_tile() 来处理此问题。看起来上面的图中可能已经进行了一些重新调整，因为没有任何负相关性，因此请在您的数据中考虑到这一点。使用 mtcars 数据集：

library(ggplot2)
library(reshape)

z <- cor(mtcars)
z.m <- melt(z)

ggplot(z.m, aes(X1, X2, fill = value)) + geom_tile() + 
scale_fill_gradient(low = "blue",  high = "yellow")

在此处输入图像描述

编辑：

ggplot(z.m, aes(X1, X2, fill = value)) + geom_tile() + 
scale_fill_gradient2(low = "blue",  high = "yellow")

< img src="https://i.sstatic.net/dpdSK.png" alt="在此处输入图像描述">

允许指定中点的颜色，默认为白色，因此这里可能是一个不错的调整。其他选项可以在 ggplot 网站此处和此处。

The ggplot2 library can handle this with geom_tile(). It looks like there may have been some rescaling done in that plot above as there aren't any negative correlations, so take that into consideration with your data. Using the mtcars dataset:

library(ggplot2)
library(reshape)

z <- cor(mtcars)
z.m <- melt(z)

ggplot(z.m, aes(X1, X2, fill = value)) + geom_tile() + 
scale_fill_gradient(low = "blue",  high = "yellow")

enter image description here

EDIT:

ggplot(z.m, aes(X1, X2, fill = value)) + geom_tile() + 
scale_fill_gradient2(low = "blue",  high = "yellow")

enter image description here

allows to specify the colour of the midpoint and it defaults to white so may be a nice adjustment here. Other options can be found on the ggplot website here and here.

回复收藏 0 原文

江南月 2024-11-03 01:29:43

使用 corrplot 包：

library(corrplot)
data(mtcars)
M <- cor(mtcars)
##  different color series
col1 <- colorRampPalette(c("#7F0000","red","#FF7F00","yellow","white", 
        "cyan", "#007FFF", "blue","#00007F"))
col2 <- colorRampPalette(c("#67001F", "#B2182B", "#D6604D", "#F4A582", "#FDDBC7",
        "#FFFFFF", "#D1E5F0", "#92C5DE", "#4393C3", "#2166AC", "#053061"))  
col3 <- colorRampPalette(c("red", "white", "blue")) 
col4 <- colorRampPalette(c("#7F0000","red","#FF7F00","yellow","#7FFF7F", 
        "cyan", "#007FFF", "blue","#00007F"))   
wb <- c("white","black")


par(ask = TRUE)


## different color scale and methods to display corr-matrix
corrplot(M, method="number", col="black", addcolorlabel="no")
corrplot(M, method="number")
corrplot(M)
corrplot(M, order ="AOE")
corrplot(M, order ="AOE", addCoef.col="grey")

corrplot(M, order="AOE", col=col1(20), cl.length=21,addCoef.col="grey")
corrplot(M, order="AOE", col=col1(10),addCoef.col="grey")

corrplot(M, order="AOE", col=col2(200))
corrplot(M, order="AOE", col=col2(200),addCoef.col="grey")
corrplot(M, order="AOE", col=col2(20), cl.length=21,addCoef.col="grey")
corrplot(M, order="AOE", col=col2(10),addCoef.col="grey")

corrplot(M, order="AOE", col=col3(100))
corrplot(M, order="AOE", col=col3(10))



corrplot(M, method="color", col=col1(20), cl.length=21,order = "AOE", addCoef.col="grey")

if(TRUE){

corrplot(M, method="square", col=col2(200),order = "AOE")

corrplot(M, method="ellipse", col=col1(200),order = "AOE")


corrplot(M, method="shade", col=col3(20),order = "AOE")

corrplot(M, method="pie", order = "AOE")


## col=wb
corrplot(M, col = wb, order="AOE", outline=TRUE, addcolorlabel="no")
## like Chinese wiqi, suit for either on screen or white-black print.
corrplot(M, col = wb, bg="gold2",  order="AOE", addcolorlabel="no")
}

例如：

在此处输入图像描述

IMO 相当优雅

Use the corrplot package:

library(corrplot)
data(mtcars)
M <- cor(mtcars)
##  different color series
col1 <- colorRampPalette(c("#7F0000","red","#FF7F00","yellow","white", 
        "cyan", "#007FFF", "blue","#00007F"))
col2 <- colorRampPalette(c("#67001F", "#B2182B", "#D6604D", "#F4A582", "#FDDBC7",
        "#FFFFFF", "#D1E5F0", "#92C5DE", "#4393C3", "#2166AC", "#053061"))  
col3 <- colorRampPalette(c("red", "white", "blue")) 
col4 <- colorRampPalette(c("#7F0000","red","#FF7F00","yellow","#7FFF7F", 
        "cyan", "#007FFF", "blue","#00007F"))   
wb <- c("white","black")


par(ask = TRUE)


## different color scale and methods to display corr-matrix
corrplot(M, method="number", col="black", addcolorlabel="no")
corrplot(M, method="number")
corrplot(M)
corrplot(M, order ="AOE")
corrplot(M, order ="AOE", addCoef.col="grey")

corrplot(M, order="AOE", col=col1(20), cl.length=21,addCoef.col="grey")
corrplot(M, order="AOE", col=col1(10),addCoef.col="grey")

corrplot(M, order="AOE", col=col2(200))
corrplot(M, order="AOE", col=col2(200),addCoef.col="grey")
corrplot(M, order="AOE", col=col2(20), cl.length=21,addCoef.col="grey")
corrplot(M, order="AOE", col=col2(10),addCoef.col="grey")

corrplot(M, order="AOE", col=col3(100))
corrplot(M, order="AOE", col=col3(10))



corrplot(M, method="color", col=col1(20), cl.length=21,order = "AOE", addCoef.col="grey")

if(TRUE){

corrplot(M, method="square", col=col2(200),order = "AOE")

corrplot(M, method="ellipse", col=col1(200),order = "AOE")


corrplot(M, method="shade", col=col3(20),order = "AOE")

corrplot(M, method="pie", order = "AOE")


## col=wb
corrplot(M, col = wb, order="AOE", outline=TRUE, addcolorlabel="no")
## like Chinese wiqi, suit for either on screen or white-black print.
corrplot(M, col = wb, bg="gold2",  order="AOE", addcolorlabel="no")
}

For example:

enter image description here

Rather elegant IMO

回复收藏 0 原文

苏璃陌 2024-11-03 01:29:43

这种类型的图表在其他术语中被称为“热图”。获得相关矩阵后，请使用各种教程之一绘制它。

使用基础图形：
http:// Flowingdata.com/2010/01/21/how-to-make-a-heatmap-a-quick-and-easy-solution/

使用 ggplot2：
http://learnr.wordpress.com/2010/01/ 26/ggplot2-快速热图-绘图/

回复收藏 0 原文

寄居人 2024-11-03 01:29:43

我一直在研究类似于 @daroczig 发布的可视化的内容，其中 @Ulrik 使用 ellipse 包的 plotcorr() 函数发布了代码。我喜欢使用椭圆来表示相关性，以及使用颜色来表示负相关和正相关。然而，我希望引人注目的颜色能够在接近 1 和 -1 的相关性中脱颖而出，而不是在接近 0 的相关性中脱颖而出。

我创建了一种替代方案，将白色椭圆叠加在彩色圆圈上。每个白色椭圆的大小都经过调整，使其后面可见的彩色圆圈的比例等于相关性的平方。当相关性接近 1 和 -1 时，白色椭圆很小，大部分彩色圆圈可见。当相关性接近 0 时，白色椭圆很大，并且几乎看不到彩色圆圈。

函数 plotcor() 位于 https://github.com/JVAdams/jvamisc/blob/master/R/plotcor.r。

使用 mtcars 数据集生成的绘图示例如下所示。

library(plotrix)
library(seriation)
library(MASS)
plotcor(cor(mtcars), mar=c(0.1, 4, 4, 0.1))

调用plotcor()函数的结果

I have been working on something similar to the visualization posted by @daroczig, with code posted by @Ulrik using the plotcorr() function of the ellipse package. I like the use of ellipses to represent correlations, and the use of colors to represent negative and positive correlation. However, I wanted the eye-catching colors to stand out for correlations close to 1 and -1, not for those close to 0.

I created an alternative in which white ellipses are overlaid on colored circles. Each white ellipse is sized so that the proportion of the colored circle visible behind it is equal to the squared correlation. When the correlation is near 1 and -1, the white ellipse is small, and much of the colored circle is visible. When the correlation is near 0, the white ellipse is large, and little of the colored circle is visible.

The function, plotcor(), is available at https://github.com/JVAdams/jvamisc/blob/master/R/plotcor.r.

An example of the resulting plot using the mtcars dataset is shown below.

library(plotrix)
library(seriation)
library(MASS)
plotcor(cor(mtcars), mar=c(0.1, 4, 4, 0.1))

result of call to plotcor() function

回复收藏 0 原文

淡墨 2024-11-03 01:29:43

corrplot R 包中的 corrplot() 函数也可用于绘制相关图。

library(corrplot)  
M<-cor(mtcars) # compute correlation matrix
corrplot(M, method="circle")

这里发布了几篇描述如何计算和可视化相关矩阵的文章：

The corrplot() function from corrplot R package can be also used to plot a correlogram.

library(corrplot)  
M<-cor(mtcars) # compute correlation matrix
corrplot(M, method="circle")

several articles describing how to compute and visualize correlation matrix are published here:

回复收藏 0 原文

分分钟 2024-11-03 01:29:43

我意识到已经有一段时间了，但新读者可能会对 corrr 包中的 rplot() 感兴趣 (https://cran.rstudio.com/web/packages/corrr/index.html)，它可以产生各种绘制@daroczig提到的，但设计数据管道方法：

install.packages("corrr")
library(corrr)
mtcars %>% correlate() %>% rplot()

mtcars %>% correlate() %>% rearrange() %>% rplot()

mtcars %>% correlate() %>% rearrange() %>% rplot(shape = 15)

mtcars %>% correlate() %>% rearrange() %>% shave() %>% rplot(shape = 15)

mtcars %>% correlate() %>% rearrange(absolute = FALSE) %>% rplot(shape = 15)

在此处输入图像描述 < /a>

I realise that it's been a while, but new readers might be interested in rplot() from the corrr package (https://cran.rstudio.com/web/packages/corrr/index.html), which can produce the sorts of plots @daroczig mentions, but design for a data pipeline approach:

install.packages("corrr")
library(corrr)
mtcars %>% correlate() %>% rplot()

mtcars %>% correlate() %>% rearrange() %>% rplot()

mtcars %>% correlate() %>% rearrange() %>% rplot(shape = 15)

mtcars %>% correlate() %>% rearrange() %>% shave() %>% rplot(shape = 15)

mtcars %>% correlate() %>% rearrange(absolute = FALSE) %>% rplot(shape = 15)

回复收藏 0 原文

月光色 2024-11-03 01:29:43

我最近了解到的另一个解决方案是使用 qtlcharts 包创建的交互式热图。

install.packages("qtlcharts")
library(qtlcharts)
iplotCorr(mat=mtcars, group=mtcars$cyl, reorder=TRUE)

下面是结果图的静态图像。

您可以在我的博客。将鼠标悬停在热图上可查看行、列和单元格值。单击单元格可查看带有按组着色的符号的散点图（在此示例中，为气缸数，4 为红色，6 为绿色，8 为蓝色）。将鼠标悬停在散点图中的点上会给出行的名称（在本例中为汽车的品牌）。

Another solution I recently learned about is an interactive heatmap created with the qtlcharts package.

install.packages("qtlcharts")
library(qtlcharts)
iplotCorr(mat=mtcars, group=mtcars$cyl, reorder=TRUE)

Below is a static image of the resulting plot.

You can see the interactive version on my blog. Hover over the heatmap to see the row, column, and cell values. Click on a cell to see a scatterplot with symbols colored by group (in this example, the number of cylinders, 4 is red, 6 is green, and 8 is blue). Hovering over the points in the scatterplot gives the name of the row (in this case the make of the car).

回复收藏 0 原文

谈下烟灰 2024-11-03 01:29:43

另一种选择是将 GGally 包与 ggcorr 函数如下：

library(GGally)
ggcorr(mtcars, method = c("everything", "pearson"), label = TRUE)

ggcorr(mtcars, method = c("everything", "pearson"), label = TRUE, geom = "circle")

^{创建于 2022 年 8 月 20 日，使用 reprex v2.0.2}

检查上面的链接以获取更多选项。

Another option is using the GGally package with ggcorr function like this:

library(GGally)
ggcorr(mtcars, method = c("everything", "pearson"), label = TRUE)

ggcorr(mtcars, method = c("everything", "pearson"), label = TRUE, geom = "circle")

^{Created on 2022-08-20 with reprex v2.0.2}

Check the links above for a lot of more options.

回复收藏 0 原文

傾旎 2024-11-03 01:29:43

因为我无法发表评论，所以我必须将 2c 给出 daroczig 的答案作为 anwser...

椭圆散点图确实来自 ellipse 包并使用以下命令生成：（

corr.mtcars <- cor(mtcars)
ord <- order(corr.mtcars[1,])
xc <- corr.mtcars[ord, ord]
colors <- c("#A50F15","#DE2D26","#FB6A4A","#FCAE91","#FEE5D9","white",
            "#EFF3FF","#BDD7E7","#6BAED6","#3182BD","#08519C")   
plotcorr(xc, col=colors[5*xc + 6])

来自手册页）

corrplot 包也可能 - 如建议的那样- 对漂亮的图像很有用在这里找到

Since I cannot comment, I have to give my 2c to the answer by daroczig as an anwser...

The ellipse scatter plot is indeed from the ellipse package and generated with:

corr.mtcars <- cor(mtcars)
ord <- order(corr.mtcars[1,])
xc <- corr.mtcars[ord, ord]
colors <- c("#A50F15","#DE2D26","#FB6A4A","#FCAE91","#FEE5D9","white",
            "#EFF3FF","#BDD7E7","#6BAED6","#3182BD","#08519C")   
plotcorr(xc, col=colors[5*xc + 6])

(from the man page)

The corrplot package may also - as suggested - be useful with pretty images found here

回复收藏 0 原文

花间憩 2024-11-03 01:29:43

这是层次聚类热图（带有树状图）的教科书示例。使用 gplots heatmap.2 因为它优于基本热图，但想法是相同的。 colorRampPalette 帮助生成 50 种（过渡）颜色。

library(gplots)

heatmap.2(cor(mtcars), trace="none", col=colorRampPalette(c("blue2","white","red3"))(50))

This is a textbook example for a hierarchical clustering heatmap (with dendrogram). Using gplots heatmap.2 because it's superior to the base heatmap, but the idea is the same. colorRampPalette helps generating 50 (transitional) colors.

library(gplots)

heatmap.2(cor(mtcars), trace="none", col=colorRampPalette(c("blue2","white","red3"))(50))

回复收藏 0 原文

~没有更多了~

关于作者

梦初启

暂无简介

0 文章

0 评论

24 人气

关注发私信

友情链接

文江博客

如何将相关矩阵绘制成图表？

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

评论（13）

关于作者

相关话题

热门标签

推荐作者

Gabu-gabumon

qq_CgiN62

荔枝明

赏烟花じ飞满天

独守阴晴ぅ圆缺

¤→小豸慧

友情链接

如何将相关矩阵绘制成图表？

如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

评论（13）

关于作者

相关话题

热门标签

推荐作者

Gabu-gabumon

qq_CgiN62

荔枝明

赏烟花じ飞满天

独守阴晴ぅ圆缺

¤→小豸慧

友情链接

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。