https://communities.sas.com/t5/SAS-Procedures/How-can-I-specify-a-prior-in-proc-fmm-using-the-Bayes-statement/m-p/268250#M58079 <P>I want to estimate a univariate Gaussian mixture model with two components using proc fmm. With respect to the Bayes statement, I am assuming a triangular distribution for one parameter. Does anybody know how to implement it? I checked proc mcmc and there it is possible to construct your own prior via the general function (<A href="https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_mcmc_sect026.htm)." target="_blank">https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_mcmc_sect026.htm).</A> <BR /><BR />Many thanks in advance.</P> Wed, 04 May 2016 11:52:18 GMT DaLack 2016-05-04T11:52:18Z https://communities.sas.com/t5/SAS-Procedures/How-can-I-specify-a-prior-in-proc-fmm-using-the-Bayes-statement/m-p/268250#M58079 <P>I want to estimate a univariate Gaussian mixture model with two components using proc fmm. With respect to the Bayes statement, I am assuming a triangular distribution for one parameter. Does anybody know how to implement it? I checked proc mcmc and there it is possible to construct your own prior via the general function (<A href="https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_mcmc_sect026.htm)." target="_blank">https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_mcmc_sect026.htm).</A> <BR /><BR />Many thanks in advance.</P> Wed, 04 May 2016 11:52:18 GMT https://communities.sas.com/t5/SAS-Procedures/How-can-I-specify-a-prior-in-proc-fmm-using-the-Bayes-statement/m-p/268250#M58079 DaLack 2016-05-04T11:52:18Z https://communities.sas.com/t5/SAS-Procedures/How-can-I-specify-a-prior-in-proc-fmm-using-the-Bayes-statement/m-p/268292#M58080 <P>I don't think that FMM supports general priors. I only see the "usual suspects" in the doc: beta, gama, Dirichlet, etc</P> <P>&nbsp;</P> <P>However, there is an easy way to create a beta distribution that kinda-sorta looks like a triangular distribution, in the sense that it has support is on [a,b] and it has a mode at c. &nbsp;It is sometimes known as the PERT distribution. See the article <A href="http://blogs.sas.com/content/iml/2012/10/24/pert-distribution.html" target="_self">"That distribution is quite PERT!"</A>&nbsp;to learn how to choose the parameters of the beta distribution. &nbsp;Choosing a PERT (via beta) distribution might capture the spirit of your&nbsp;prior.</P> Wed, 04 May 2016 14:37:53 GMT https://communities.sas.com/t5/SAS-Procedures/How-can-I-specify-a-prior-in-proc-fmm-using-the-Bayes-statement/m-p/268292#M58080 Rick_SAS 2016-05-04T14:37:53Z