PDA

View Full Version : Generating Normal and Nonnormal Multivariate Mixture Data



Havilah
07-03-2012, 12:04 AM
Hello All,
I am searching for a consultant to help me with my dissertation.

I am doing a Monte Carlo simulation study that compares the ability of four different methods to detect clusters in data (factor mixture modeling, Q factor analysis, and two other methods). I am testing the methods on different data conditions: sample size, distributional assumption of variables, mixture proportions, number of clusters, etc. I used Fan’s method outlined in his text but something is off. I have spent way too much time on data generation, considering it is not the overarching research question.

I used SAS for the data simulations. But am open to SPSS, R or Mplus.

Is there anyone out there that has done such in the past and would like to be hired to do such?