在R语言中创建矩阵，是否有更高效的方法？

Great question—manually typing every element of a covariance matrix is super error-prone, and there’s a much cleaner, statistically intuitive way to build this using the core properties of covariance matrices.

The Key Idea #

A covariance matrix can be constructed from two simpler, easier-to-manage components:

1. A vector of standard deviations for each variable
2. A correlation matrix that defines pairwise correlations between variables

Mathematically, if sd_vec is your standard deviation vector and cor_mat is your correlation matrix, the covariance matrix sig is calculated as:
sig = diag(sd_vec) %*% cor_mat %*% diag(sd_vec)

Step-by-Step Implementation #

Let’s apply this to your specific case:

1. Define the standard deviation vector
   From your original matrix, the standard deviations of the four variables are clearly 0.56, 7.77, 13.48, and 16.64:

   sd_vec <- c(0.56, 7.77, 13.48, 16.64)
   

2. Define the correlation matrix
   Extract the pairwise correlation values from your original calculation (remember, correlation matrices are symmetric, with 1s on the diagonal):

   cor_mat <- matrix(
     c(1,      -0.07,  -0.095, -0.095,
       -0.07,  1,      0.959,  0.936,
       -0.095, 0.959,  1,      0.997,
       -0.095, 0.936,  0.997,  1),
     nrow = 4, ncol = 4, byrow = TRUE
   )
   

3. Compute the covariance matrix
   Use matrix multiplication to combine the two components automatically:

   sig <- diag(sd_vec) %*% cor_mat %*% diag(sd_vec)
   

Why This Is Better #

• Less error-prone: No need to manually compute tedious cross-terms like 0.56*7.77*-0.07—the math is handled reliably by R
• More readable: The code directly maps to statistical concepts, so anyone reading it can immediately understand what’s being built
• Easier to modify: If you need to adjust a standard deviation or correlation, you only change one value instead of hunting down all related manual calculations

You can verify this produces the exact same matrix as your manual input by running:

# Your original manual matrix
original_sig <- matrix(c(0.56^2, 0.56*7.77*-0.07, 0.56*13.48*-0.095, 0.56*16.64*-0.095, 0.56*7.77*-0.07, 7.77^2, 7.77*13.48*0.959, 7.77*16.64*0.936, 0.56*13.48*-0.095, 7.77*13.48*0.959, 13.48^2, 13.48*16.64*0.997, 0.56*16.64*-0.095, 7.77*16.64*0.936, 13.48*16.64*0.997, 16.64^2), nrow = 4, ncol = 4)

# Check if they're identical
all.equal(sig, original_sig)

内容的提问来源于stack exchange，提问作者Alex