topic General number of components in Statistical Procedures

General number of components

Jess03ng — Fri, 20 Nov 2020 09:58:42 GMT

First, I have five variables used for analysis. How to compute a general number of components analysis in proc fmm which comprises of these five variables?

Re: General number of components

PGStats — Fri, 20 Nov 2020 19:38:23 GMT

From the proc fmm doc.

Basic Features

The FMM procedure estimates the parameters in univariate finite mixture models and produces various statistics to evaluate parameters and model fit.

Maybe you are looking for clustering procedures?

If you are trying to estimate the number of clusters in your data, read this discussion on the topic.

Re: General number of components

Rick_SAS — Fri, 20 Nov 2020 21:21:36 GMT

To expand PGStats's comment, PROC FMM is a regression procedure in which you model a single response variable (Y) as a finite mixture of distributions that are linear functions of the covariates. It sounds like you want a clustering algorithm for data in 5-dimensional space. If you show sample data and explain more, we can provide more suggestions.