Results 1 to 2 of 2

Math Help - Estimate paramters of gaussian likelihood and gaussian prior from posterior sample?

  1. #1
    Junior Member
    Joined
    Dec 2010
    Posts
    62

    Estimate paramters of gaussian likelihood and gaussian prior from posterior sample?

    Hi,
    I have a bunch of 2D data which I want to interpret as samples from 11 different posterior (gaussian) distributions. I can obviously estimate the covariance and mean of these posteriors from the data, but what I really want to do is estimate the parameters of the underlying (gaussian) likelihood function and (gaussian) prior distribution.

    Can anyone tell me if this is possible, and if so how, or point me to a source?

    Many thanks, MD
    Follow Math Help Forum on Facebook and Google+

  2. #2
    MHF Contributor
    Joined
    Nov 2013
    From
    California
    Posts
    2,475
    Thanks
    956

    Re: Estimate paramters of gaussian likelihood and gaussian prior from posterior sampl

    It sounds like you want to do a cluster analysis to estimate which points belong to which underlying distribution, then having decided on which samples belong to which distribution you can easily find the sample mean and variance of each of the clusters.

    Is this along the lines of what you are looking for? You can google Cluster Analysis and it should point in the right direction.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Prior and Posterior Distributions
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: October 14th 2010, 06:32 AM
  2. Prior/Posterior Distribution
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: October 5th 2009, 10:59 PM
  3. Prior and posterior distributions
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: August 13th 2009, 12:42 PM
  4. Replies: 0
    Last Post: March 30th 2008, 12:44 PM
  5. Replies: 2
    Last Post: March 25th 2008, 07:38 AM

Search Tags


/mathhelpforum @mathhelpforum