Skip to main content
added 69 characters in body
Source Link
Glen_b
  • 290.5k
  • 37
  • 652
  • 1.1k

A covariance matrix - or more properly, a variance-covariance matrix, consists of variance terms on the diagonal and covariances off the diagonal. It is also symmetric.

Hence if the matrix you have is actually a 2x2 covariance matrix, either the (1,2) or the (2,1) element is the covariance of the two variables (they'll be identical). The other two elements are the two variances.

(Which is which depends on the order of the variables.)

Edit: Out of curiosity, how were you computing the covariance matrix?

A covariance matrix - or more properly, a variance-covariance matrix, consists of variance terms on the diagonal and covariances off the diagonal. It is also symmetric.

Hence if the matrix you have is actually a 2x2 covariance matrix, either the (1,2) or the (2,1) element is the covariance of the two variables (they'll be identical). The other two elements are the two variances.

(Which is which depends on the order of the variables.)

A covariance matrix - or more properly, a variance-covariance matrix, consists of variance terms on the diagonal and covariances off the diagonal. It is also symmetric.

Hence if the matrix you have is actually a 2x2 covariance matrix, either the (1,2) or the (2,1) element is the covariance of the two variables (they'll be identical). The other two elements are the two variances.

(Which is which depends on the order of the variables.)

Edit: Out of curiosity, how were you computing the covariance matrix?

Source Link
Glen_b
  • 290.5k
  • 37
  • 652
  • 1.1k

A covariance matrix - or more properly, a variance-covariance matrix, consists of variance terms on the diagonal and covariances off the diagonal. It is also symmetric.

Hence if the matrix you have is actually a 2x2 covariance matrix, either the (1,2) or the (2,1) element is the covariance of the two variables (they'll be identical). The other two elements are the two variances.

(Which is which depends on the order of the variables.)