1

Possible Duplicate:
Reducing no of variables subsetted based on depth for PCA

I have a question, I am trying to apply a method to my research area, which has not been aplied yet, based on publications and literature I could find.

The case is the following:

I have a dataset with x variables and these variables are changing with depth for every observation. I will show a simple dataset:

sampledata

Sorry I had no idea how to enter a table so I eneterd a pic. It is important how the curve looks like, so it would be nice to cluster the observations based on their curves(characteristic), which reoresents the x variables.

My questions are:

What would be the most appropriate method? Can I reduce the number of curves with PCA? Can I combine the curves of the x variables to one? Or is there a method, which can cluster my observations based on their x curves?

I'd prefer an R or Python solution, because I have no acccess to matlab.

Thank You in advance

-v

Community
  • 1
user1775772
  • 99
  • 1
  • 4
  • possible duplicate of [Reducing no of variables subsetted based on depth for PCA](http://stats.stackexchange.com/questions/43529/reducing-no-of-variables-subsetted-based-on-depth-for-pca) OR [Is it possible to do time-series clustering based on curve shape?](http://stats.stackexchange.com/questions/3331/is-it-possible-to-do-time-series-clustering-based-on-curve-shape) – Has QUIT--Anony-Mousse Nov 18 '12 at 08:57

0 Answers0