+ Reply to Thread
Results 1 to 3 of 3

Thread: Example of PDF with correlation of a specific value

  1. #1
    Points: 1,805, Level: 25
    Level completed: 5%, Points required for next Level: 95

    Posts
    9
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Example of PDF with correlation of a specific value




    For example example of a pdf or pmf with correlation . It is obviously easy to find ones with 0,1,-1. But what would be a good general way to compute one with a specific value?
    Last edited by darkdream; 11-13-2013 at 12:36 AM.

  2. #2
    Devorador de queso
    Points: 95,819, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,935
    Thanks
    307
    Thanked 2,629 Times in 2,245 Posts

    Re: Example of PDF with correlation of a specific value

    You're asking a very general question. If your goal is just to generate correlated observations with a set correlation you could look into the bivariate normal distribution.
    I don't have emotions and sometimes that makes me very sad.

  3. The Following User Says Thank You to Dason For This Useful Post:

    darkdream (11-11-2013)

  4. #3
    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: Example of PDF with correlation of a specific value


    It seems that you are having a set of independent random variables and want a copula to give them a specific correlation structure.

    Dason has pointed out a common example - the multivariate normal copula which uses the covariance matrix as an input of parameter. In general this will not give you the answer directly as it involve some nonlinear transformation (probability integral transform)

    Another very simple example is that you may consider the Cholesky decomposition of your desired correlation matrix. Then by applying the resulting linear transformation you can always achieve the desired correlation structure (with proper scaling of course). The limitation is that it is a linear transformation of the random variable - anyway it should be a trivial example to start with.

+ Reply to 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