Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 1 down vote favorite
1

I have a training data set that I've done some cluster analysis on using R's pvclust package. I have a dendrogram of the results and am currently looking for the best way to use the information to create a model(s) of the data. Can anyone recommend a resource to walk me thru the process of going from clustering to predictive model? In other words, how do I read a dendrogram to make the most appropriate model(s).

share|improve this question
Some clarification might be helpful. You say want to build a 'predictive model', what is it you want to predict? Also, what constitute the data that went into the CA? A set of explanatory variables, a set of response variables, all the variables you had access to, etc. – gung Nov 18 '11 at 22:09
The training data was missing the variable I wanted to predict (a credit event/failure) and I ran the CA on everything else (age, income, lateness, etc.) The dendrogram normalized the data so that it is on a 0-3 scale. I'm trying to understand what the dendrogram means. I basically want to create a model that better predicts whether a person will have a credit failure. – screechOwl Nov 18 '11 at 22:50

1 Answer

up vote 2 down vote accepted

One way to proceed is to put the dummies of the clusters & run the regression. For example if there are n cluster groups, then (n-1) cluster dummies & run the regression using these dummies.

share|improve this answer
Do you have a link that would describe how to do this in detail? – screechOwl Nov 21 '11 at 18:23
Sorry Acesnap I don't have the link. But it's pretty easy to do. – Ari Nov 21 '11 at 21:27

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.