r - geom_density_2d 中的轮廓太紧

Question

我希望轮廓一直围绕数据，但我的密度区域很小。有谁知道如何解决这个问题？我将附上我的代码和数据。我尝试了不同的参数，但我似乎无法改变它。我想我想要的一种思考方式是 100% 的轮廓，而不是 90% 的轮廓，然后是 80% 等等。

ggplot(df3, aes(x, y)) +
stat_density_2d(size = 1, aes(color = "Contours")) +
geom_path(data = df3, aes(color = "Parameter Path"), lwd = 1) +
geom_point(data = df3, aes(color = "Accepted Values"), size = 2) +
labs(x = expression(mu), y = expression(phi)) +
theme(axis.text.x = element_text(size = 18),
    axis.title.x = element_text(size = 20),
    axis.text.y = element_text(size = 18),
    axis.title.y = element_text(size = 20, angle = 0, vjust = 0.5),
    plot.title = element_text(size = 20, face = "bold", hjust = 0.5),
    legend.position = c(0.14, 0.84),
    legend.background = element_rect(fill = "grey96", size = 1, linetype = 
"solid"),
    legend.text = element_text(size = 16),
    legend.title = element_text(size = 18, face = "bold")) +
guides(fill = guide_legend(reverse = TRUE), 
color=guide_legend(override.aes=list(fill=NA))) +
scale_x_continuous(breaks = seq(0.065, 0.135, 0.01),
                 limits = c(0.065, 0.135)) +
scale_y_continuous(breaks = seq(0.03, 0.055, 0.005),
                 limits = c(0.03, 0.055)) +
ggtitle("Confidence Ellipse") +
scale_color_manual("Results", values = c("red1", "deepskyblue2", "black"), 
breaks = c("Accepted Values", "Parameter Path", "Contours"))

数据：

> x0
 [1] 0.08358176 0.06553959 0.08701262 0.13261197 0.11827522 0.12895425         
     0.07641320 0.09780957 0.12004418 0.10330265 0.13115285 0.11239738
[13] 0.09539895 0.09978812 0.08133239 0.08256920 0.06962943 0.11592930 
     0.06910299 0.08968366 0.09268526 0.09354692 0.09407394 0.10241154
[25] 0.11241420 0.10679109 0.10047328 0.08391576 0.09663064 0.09601004 
     0.08067915 0.09624377 0.10755340 0.09217993 0.10787710 0.08061391
[37] 0.08068861 0.08953676 0.09189470 0.12449563 0.09883104 0.12972641 
     0.11940337 0.11907046 0.10960429 0.09448710

> y0
 [1] 0.04219829 0.03586581 0.03353818 0.05380570 0.05151715 0.05587795 
     0.03240938 0.03869719 0.04355228 0.03748867 0.05029841 0.04462947
[13] 0.04057039 0.03691092 0.03801298 0.03997664 0.03429455 0.04798272 
     0.03641225 0.04115275 0.04029572 0.04043962 0.03793742 0.03940454
[25] 0.04468192 0.04189995 0.04296391 0.03632054 0.04192571 0.03698961 
     0.03703587 0.04008437 0.04824601 0.03933985 0.04131714 0.03589466
[37] 0.03665176 0.03872114 0.04056889 0.04843315 0.04547021 0.04853519 
     0.04710753 0.04888831 0.04833231 0.03744968

df3 = data.frame(x = x0, y = y0)

标签： rggplot2

Answer 1

如果我在没有任何更改的情况下运行您的代码，我会得到下面的图，我认为这就是您所追求的。

我按以下方式导入上面的数据：

library(dplyr)
library(ggplot2)

y0 <- t(read.table(text = "0.04219829 0.03586581 0.03353818 0.05380570 0.05151715 0.05587795 0.03240938 0.03869719 0.04355228 0.03748867 0.05029841 0.04462947 0.04057039 0.03691092 0.03801298 0.03997664 0.03429455 0.04798272 0.03641225 0.04115275 0.04029572 0.04043962 0.03793742 0.03940454 0.04468192 0.04189995 0.04296391 0.03632054 0.04192571 0.03698961 0.03703587 0.04008437 0.04824601 0.03933985 0.04131714 0.03589466 0.03665176 0.03872114 0.04056889 0.04843315 0.04547021 0.04853519 0.04710753 0.04888831 0.04833231 0.03744968",
                 sep=" ",header=FALSE,dec="."))


x0 <- t(read.table(text = "0.08358176 0.06553959 0.08701262 0.13261197 0.11827522 0.12895425 0.07641320 0.09780957 0.12004418 0.10330265 0.13115285 0.11239738 0.09539895 0.09978812 0.08133239 0.08256920 0.06962943 0.11592930 0.06910299 0.08968366 0.09268526 0.09354692 0.09407394 0.10241154 0.11241420 0.10679109 0.10047328 0.08391576 0.09663064 0.09601004 0.08067915 0.09624377 0.10755340 0.09217993 0.10787710 0.08061391 0.08068861 0.08953676 0.09189470 0.12449563 0.09883104 0.12972641 0.11940337 0.11907046 0.10960429 0.09448710",
                   sep=" ",header=FALSE,dec="."))

df3 = data.frame(x = x0, y = y0)

我运行您的代码（没有任何更改）

我得到以下warning messages

Warning messages:
1: Removed 1 rows containing non-finite values (stat_density2d). 
2: Removed 1 rows containing missing values (geom_point). 

可能是版本问题：

version.string R version 3.5.3 (2019-03-11)
> packageVersion("ggplot2")
[1] ‘3.1.1’

如果我使用你的版本运行ggplot2

> packageVersion("ggplot2")
[1] ‘3.0.0’

我仍然使用相同的代码得到与上面相同的情节。

此外，您可以geom_density_2d不加修饰地尝试，看看是否至少可以按预期生成轮廓：

ggplot(df3, aes(x, y)) +
  geom_density_2d(size=1,aes(color = "Contours"))+
  geom_point(data = df3, aes(color = "Accepted Values"), size = 2)