Skip to main content
Cross Validated

Return to Question

added 3 characters in body
Source Link
Lerner Zhang
Lerner Zhang
  • 6.9k
  • 2
  • 44
  • 81

To the best of my knowledge, mutual information (MI) is zero if and only if the variables are independent. I have simulated copula data and computed the MI and the results are as follows:

  1. Independent copula (MI = 0.04)
  2. Very weak dependency Gaussian copula (MI = -0.03)!

  • My question is, is it acceptable to have negative MI? isIs that because the relationship between mutual information and copula is negative?

  • Is it acceptable to have > 0 for completely independent variables.

I used infotheoan 'infotheo` package in R.

To the best of my knowledge, mutual information (MI) is zero if and only if the variables are independent. I have simulated copula data and computed the MI and the results are as follows:

  1. Independent copula (MI = 0.04)
  2. Very weak dependency Gaussian copula (MI = -0.03)!

  • My question is, is it acceptable to have negative MI? is that because the relationship between mutual information and copula is negative?

  • Is it acceptable to have > 0 for completely independent variables.

I used infotheo package in R.

To the best of my knowledge, mutual information (MI) is zero if and only if the variables are independent. I have simulated copula data and computed the MI and the results are as follows:

  1. Independent copula (MI = 0.04)
  2. Very weak dependency Gaussian copula (MI = -0.03)!

  • My question is, is it acceptable to have negative MI? Is that because the relationship between mutual information and copula is negative?

  • Is it acceptable to have > 0 for completely independent variables.

I used an 'infotheo` package in R.

edited tags; edited title
Link
Richard Hardy
Richard Hardy
  • 69.5k
  • 13
  • 126
  • 278

mutual Mutual information is not zero for compleletly independent variables and negative for weakweakly dependency

added 1 character in body
Source Link
Maryam
Maryam
  • 1.7k
  • 1
  • 9
  • 27

To the best of my knowledge, mutual information (MI) is zero if and only if the variables are independent. I have simulated copula data and computed the MI and the results are as follows:

  1. Independent copula (MI = 0.04)
  2. Very weak dependency Gaussian copula (MI = -0.03)!

  • My question is, is it acceptable to have negative MI? is that because the relationship between mutual information and copula is negative?

  • Is it acceptable to have > 0 for completely independent variables.

I used ifotheoinfotheo package in R.

To the best of my knowledge, mutual information (MI) is zero if and only if the variables are independent. I have simulated copula data and computed the MI and the results are as follows:

  1. Independent copula (MI = 0.04)
  2. Very weak dependency Gaussian copula (MI = -0.03)!

  • My question is, is it acceptable to have negative MI? is that because the relationship between mutual information and copula is negative?

  • Is it acceptable to have > 0 for completely independent variables.

I used ifotheo package in R.

To the best of my knowledge, mutual information (MI) is zero if and only if the variables are independent. I have simulated copula data and computed the MI and the results are as follows:

  1. Independent copula (MI = 0.04)
  2. Very weak dependency Gaussian copula (MI = -0.03)!

  • My question is, is it acceptable to have negative MI? is that because the relationship between mutual information and copula is negative?

  • Is it acceptable to have > 0 for completely independent variables.

I used infotheo package in R.

deleted 32 characters in body
Source Link
Maryam
Maryam
  • 1.7k
  • 1
  • 9
  • 27
Source Link
Maryam
Maryam
  • 1.7k
  • 1
  • 9
  • 27