+ Reply to Thread
Results 1 to 4 of 4

Thread: How to generate two random vectors correlated to three existing vectors

  1. #1
    Points: 1,359, Level: 20
    Level completed: 59%, Points required for next Level: 41

    Location
    Edinburgh
    Posts
    11
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Question How to generate two random vectors correlated to three existing vectors




    Let's have three vectors that are given and cannot change (M, N and O). I have to generate two random vectors X and Y. The correlation that has to be kept is given by the correlation matrix C. The correlation matrix show the correlation between all the vectors (M, N, O , X and Y).

    Does anyone know, how to achive this in general? Some kind of SAS code would help as well.

  2. #2
    TS Contributor
    Points: 22,410, Level: 93
    Level completed: 6%, Points required for next Level: 940

    Posts
    3,020
    Thanks
    12
    Thanked 565 Times in 537 Posts

    Re: How to generate two random vectors correlated to three existing vectors

    Maybe you are talking about

    http://en.wikipedia.org/wiki/Partial_correlation

    But in general you will need to find the conditional joint pdf for the generation.

  3. #3
    Points: 1,359, Level: 20
    Level completed: 59%, Points required for next Level: 41

    Location
    Edinburgh
    Posts
    11
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: How to generate two random vectors correlated to three existing vectors

    Actually no. Partial correlation is a theory for completely different things. It does not help generating new random variables.

  4. #4
    TS Contributor
    Points: 22,410, Level: 93
    Level completed: 6%, Points required for next Level: 940

    Posts
    3,020
    Thanks
    12
    Thanked 565 Times in 537 Posts

    Re: How to generate two random vectors correlated to three existing vectors


    So as said before do you know / can you specify the conditional joint pdf

    f_{X, Y|M, N, O}(x, y|m, n, o)

    And can it be completely characterized by a given correlation matrix (i.e. is it the only parameter of the distribution)?

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats