If I want to generate multivariate Gaussian scale mixture samples with given location and covariance (characteristic matrix) of sample how will I generate that with any mixture distribution?

For example if I want to generate samples of Gaussian scale mixture with mean 0 and Covariance R let the mixture distribution be exponential or Laplace. How to generate that?

  • $\begingroup$ You'd also need to specify the mixing distribution on the scale parameter. $\endgroup$
    – Glen_b
    Sep 28, 2015 at 12:40
  • 1
    $\begingroup$ Please clarify your question, carefully defining what you mean by each of the terms you are using. If this is for some class, add the self-study tag and modify your question to follow the guidelines there. $\endgroup$
    – Glen_b
    Sep 29, 2015 at 6:46
  • $\begingroup$ I hope now I am clear with question. It is not for some class or book. I need to do some simulations for that I need to include Gaussian scale samples as my noise. For that I want to generate the samples. $\endgroup$
    – undefined
    Sep 29, 2015 at 7:42
  • $\begingroup$ How does this problem arise? $\endgroup$
    – Glen_b
    Sep 29, 2015 at 8:13
  • 1
    $\begingroup$ I find much of your previous comment unintelligible. $\endgroup$
    – Glen_b
    Sep 29, 2015 at 10:23


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.