Gibbs Sampler

I have a dataset that is composed of data points from three clusters. Suppose the data points in the mth (m = 1,2,3) cluster follow iid N(μm,σm2 Ip) where p is the dimensionality of the data. Each data point has two dimension. (x1, y1)

I want to Implement a DA algorithm (Gibbs sampler) to cluster these data points into three groups. Can anyone help me to implement that .

Thanks so much.

[Merged information from duplicate question]

Hi,

I have a dataset that is from three clusters. Each cluster is from iid N(μm,σm2 Ip). I want to implement a DA algorithm (Gibbs sampler) to cluster these data points into three groups. Each data point has two components ( x1 ,y1) .

I would greatly appreciate if you can help me on this.

Regards,

Katere

Answer 1

[Tom Lane]

You may want to look at "help gmdistribution.fit" if you have the Statistics Toolbox.