Correaltion matrix using knitr package

Question

Is it possible to create a correlation matrix like in the image below using knitr() or similar?

for

res <- cor(iris[,1:4])

Answer 1

The scipub package has a function, correltable, to produce correlation tables with significance stars. You need to install the htmlTable package to have the possibility to get these correlation tables in HTML.

Is it what you want?

---
title: "Correlation table"
author: "Stéphane Laurent"
date: "17/01/2022"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(scipub)
library(mvtnorm)
set.seed(666L)
```

```{r}
dat <- as.data.frame(rmvnorm(50, sigma = toeplitz(3:1)))
colnames(dat) <- c("V1", "V2", "V3")
```

```{r, results='asis'}
correltable(dat, html = TRUE)
```

---

EDIT

To add the means and the standard deviations:

---
title: "Correlation table"
author: "Stéphane Laurent"
date: "17/01/2022"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(scipub)
library(htmlTable)
library(magrittr)
library(mvtnorm)
set.seed(666L)
```

```{r}
dat <- as.data.frame(rmvnorm(50, sigma = toeplitz(3:1)))
colnames(dat) <- c("V1", "V2", "V3")
cortable <- correltable(dat, html = FALSE)
table2 <- as.data.frame(matrix(as.character(cortable$table), nrow = 3)) 
caption <- cortable[["caption"]]
Means <- formatC(colMeans(dat))
Sds <- formatC(apply(dat, 2L, sd))
table1 <- data.frame(Mean = Means, SD = Sds)
```

```{r, results='asis'}
css.cell <- matrix("padding: 5px;", ncol = 6L, nrow = 4L)
css.cell[, 1L] <- 
  paste(css.cell[, 1L], "font-weight: bold;") # <-- bold row names
cbind(table1, table2) %>%
  addHtmlTableStyle(css.cell = css.cell) %>% 
  htmlTable(caption = caption)
```