[问答] 如何在相关矩阵图中显示相关系数和显著性（用星号表示）？ [推广有奖]

11楼

刺客666 发表于 2017-7-7 00:30:22

Error in corrgram(spe, order = TRUE, lower.panel = panel.shade, upper.panel = panel.conf, :
unused arguments (order = TRUE, lower.panel = panel.shade, upper.panel = panel.conf, text.panel = panel.txt)

12楼

刺客666 发表于 2017-7-7 09:35:18

你好，这个代码总是Error in corrgram(spe, order = TRUE, lower.panel = panel.shade, upper.panel = panel.conf, :
unused arguments (order = TRUE, lower.panel = panel.shade, upper.panel = panel.conf, text.panel = panel.txt)。这是怎么回事，还有就是您会在有相关系数的相关图中加入显著性的星号吗

13楼

104753170146 发表于 2019-12-30 21:03:44

cyxxxxxxxx 发表于 2016-10-5 18:39
楼主，为什么我运行完之后得到的图里面上半部分数字和星号都巨小，要怎么调整参数才能有你上传的这种效果。

library(corrgram)
panel.conf<-function (x, y, corr = NULL, col.regions, digits = 2, cex.cor,
                  ...)
{
  auto <- missing(cex.cor)
  usr <- par("usr")
  on.exit(par(usr))
  par(usr = c(0, 1, 0, 1))
  if (!is.null(corr)) {
est <- corr

est <- formatC(est*100, digits = digits, format = "d")

if (auto)
   cex.cor <- 0.7/strwidth(est)
text(0.1, 0.6, est, cex = cex.cor)
text(0.8, 0.8, Signif, cex=cex.cor, col = 2)
  }
  else {
results <- cor.test(x, y, alternative = "two.sided")
est <- results$estimate
est <- round(est, 2)
Signif <- symnum(results$p.value, corr = FALSE, na = FALSE,
                  cutpoints = c(0, 0.001, 0.01, 0.05, 0.1, 1),
                  symbols = c("***", "**", "*", ".", " "))
if (auto)
   cex.cor <- 1.5
text(0.4, 0.5, est, cex=cex.cor,col=2)

text(0.8, 0.8, Signif, cex=cex.cor, col = 2)

  }
}

corrgram(shuju1,lower.panel=panel.shade,upper.panel=panel.conf,text.panel=panel.txt,
      cex.labels=1.2,font.labels = 1, main = "相关性")
我改了下，有数字和星号了。