Correlation Matrix with Plotly — plot_corr

Computes a correlation matrix and corresponding p-values, and visualises the lower triangle as an interactive heatmap using plotly. Correlation values and significance stars can be displayed inside each tile.

Usage

plot_corr_matrix(
  data,
  method = c("pearson", "spearman"),
  use = c("pairwise.complete.obs", "complete.obs", "everything", "all.obs",
    "na.or.complete"),
  digits = 2,
  show_stars = TRUE,
  show_values = TRUE,
  colors = c("#11A7F2", "#FFFFFF", "#E00087"),
  ...
)

Arguments

data: A data frame or matrix containing only numeric variables.
method: Correlation method: "pearson" or "spearman", passed directly to stats::cor() and stats::cor.test().
use: Missing-data handling passed directly to stats::cor(). One of "pairwise.complete.obs", "complete.obs", "everything", "all.obs", or "na.or.complete".
digits: Number of decimal places for correlation labels.
show_stars: Logical; display significance stars (*, **, ***) based on p-values.
show_values: Logical; display correlation values inside tiles.
colors: A vector of colours for the heatmap colour scale. Defaults to NPDA colours.
...: Additional arguments passed to plotly::plot_ly(). These can override default plotly arguments such as showscale, or colorbar.

Value

A plotly heatmap object.

Examples

# Basic correlation matrix
plot_corr_matrix(mtcars)


# Spearman correlation without tile labels, with additional plotly arguments
plot_corr_matrix(mtcars,
            method = "spearman",
            show_values = FALSE,
            show_stars = TRUE,
            colorbar = list(title = "Correlation")) |>
plotly::layout(annotations = list(
  text = paste0("<i><b>Note:</b>",
                "*: p<0.05, **: p<0.01, ***: p<0.001.</i>"),
  x = 1, y = 1, xref = "paper", yref = "paper",
  xanchor = "right", yanchor = "top",
  showarrow = FALSE))