使用单回归线而不是多元回归线的散点图

Question

我想获得单回归线和相关系数，而不是多元回归线。我的代码如下所示。

d1 <- structure(list(x1 = c(50.42,62.59,42.53,55.38,65.23,66.17,47.20,29.40,46.22,65.91), y1 = c(7.00,7.50,4.50,7.58,4.00,4.67,3.50,3.00,3.50,4.63)), .Names = c("x1", "y1"), class = "data.frame", row.names = c(NA,-10L)) 
d2 <- structure(list(x2 = c(33.65,39.16,46.93,35.75,34.50,31.16,44.72,28.63,42.73,40.54,32.14,33.56), y2 = c(3.00,4.00,3.80,2.76,3.54,3.13,3.50,2.92,4.18,4.22,3.23,2.56)), .Names = c("x2", "y2"), class = "data.frame", row.names = c(NA, -12L)) 
library(reshape2) 
library(ggplot2) 
names(d2) <- c("x1", "y2") 
df <- rbind(melt(d1, id.vars = "x1"), melt(d2, id.vars = "x1")) 
p1 <-ggplot(df, aes(x1, y = value, colour = variable)) + 
    geom_point(size=4) + 
    theme(axis.text=element_text(size=16), 
     axis.title=element_text(size=18,face="bold"), plot.title=element_text(size=32,face="bold"), 
     legend.text=element_text(size=16), 
     legend.justification=c(0,1),legend.position=c(0,1),legend.title=element_blank() 

     ) + theme(legend.background = element_rect(
            size=0.5, linetype="solid", 
            colour ="black"), panel.border = element_rect(color="black", size=1, linetype="solid")) + theme(aspect.ratio=1) + 
    labs(x = "x axis", y = "y axis") + 
xlim(0,100) + 
ylim(0,10) + geom_smooth(method=lm, se=FALSE, linetype="solid", fullrange=TRUE) + ggtitle("My first plot") + 
    scale_colour_manual(values = c("green","red"), labels = c("d1", "d2")) 

有没有简单的方法来做到这一点？

Answer 1

在geom_smooth在aes以外添加color="black"（或任何你想要的颜色）。这将覆盖调用ggplot中指定的颜色唯美性并导致一条回归线。

geom_smooth(color="black", method="lm", se=FALSE, linetype="solid", fullrange=TRUE) + 

关于你提到的有关注释评论：d1和d2不要在df数据帧存在。试试这个：

annotate(x=20, y=5, label=paste("R = ", round(cor(df$x1, df$value),2)), geom="text", size=4)