使用 corrplot 绘制卡方结果会引发矩阵误差

Question

我正在尝试使用 corrplot 包可视化 chisq.test 的结果，但出现了一个我不明白的错误。希望有人能赐教。

library(corrplot)

no <- c(330, 470, 75, 400, 1050, 170, 90)
yes <- c(1700, 1000, 250, 2100, 500, 1250, 1200)
df <- data.frame(no, yes)

#read in fn cor.mtest
cor.mtest <- function(mat, ...) {
  mat <- as.matrix(mat)
  n <- ncol(mat)
  p.mat<- matrix(NA, n, n)
  diag(p.mat) <- 0
  for (i in 1:(n - 1)) {
    for (j in (i + 1):n) {
      tmp <- cor.test(mat[, i], mat[, j], ...)
      p.mat[i, j] <- p.mat[j, i] <- tmp$p.value
    }
  }
  colnames(p.mat) <- rownames(p.mat) <- colnames(mat)
  p.mat
}

p.mat <- cor.mtest(df)

chisq <- chisq.test(df)
M <- as.matrix(chisq$residuals)
corrplot(M, cl.pos = "b", p.mat = p.mat, sig.level = 0.005, insig = "blank")

错误是：

Error in corrplot(M, cl.pos = "b", p.mat = p.mat, sig.level = 0.005, insig = "blank") : 
The matrix is not in [-1, 1]!

似乎并不能阻止结果绘制。我想了解它在这里试图告诉我什么，但文档在这一点上没有启发性。有人可以解释吗？谢谢

Answer 1

我不认为“M”是相关矩阵，所以如果你设置is.corr = FALSE你会得到输出（虽然我不确定输出是否正确；p.mat只有 4 个值，但你在 M 中有 14 个值，所以你会收到一堆警告）：

library(corrplot)
#> corrplot 0.90 loaded

no <- c(330, 470, 75, 400, 1050, 170, 90)
yes <- c(1700, 1000, 250, 2100, 500, 1250, 1200)
df <- data.frame(no, yes)

#read in fn cor.mtest
cor.mtest <- function(mat, ...) {
  mat <- as.matrix(mat)
  n <- ncol(mat)
  p.mat<- matrix(NA, n, n)
  diag(p.mat) <- 0
  for (i in 1:(n - 1)) {
    for (j in (i + 1):n) {
      tmp <- cor.test(mat[, i], mat[, j], ...)
      p.mat[i, j] <- p.mat[j, i] <- tmp$p.value
    }
  }
  colnames(p.mat) <- rownames(p.mat) <- colnames(mat)
  p.mat
}

p.mat <- cor.mtest(df)

chisq <- chisq.test(df)
M <- as.matrix(chisq$residuals)
corrplot(M, is.corr = FALSE, cl.pos = "b", p.mat = p.mat, sig.level = 0.005, insig = "blank")
#> Warning in rownames(p.mat) == rownames(corr): longer object length is not a
#> multiple of shorter object length
#> Warning in corrplot(M, is.corr = FALSE, cl.pos = "b", p.mat = p.mat, sig.level
#> = 0.005, : p.mat and corr may be not paired, their rownames and colnames are not
#> totally same!
#> Error in data.frame(..., check.names = FALSE): arguments imply differing number of rows: 14, 4

^{由 reprex package (v2.0.1)}

于 2021-10-07 创建

在我看来，仅绘制残差图更有意义：

corrplot(M, is.corr = FALSE, cl.pos = "b", cl.length = 5, insig = "blank")

使用 corrplot 绘制卡方结果会引发矩阵误差

plotting chi-square results with corrplot raises matrix error

statistics

r

chi-squared