http://www.statmethods.net/advstats/cart.html : Could be helpful . :)

Hi,

I think one of the best options is to use the rpart (recursive partitioning) package in R. It implements most of the functionalities described in the original CART 1984 book, by Breiman, Friedman, Olshen and Stone.

The link for the package is: http://cran.r-project.org/web/packages/rpart/index.html

Best wishes

Thanks all. My worries is how can i write output= input1+input2+... in my case as i have input of order 1000. let say for example 

fit

There is no formulatoin in the form input to output for cart model. The output you will obtain is a set of subgroups of the initial sample, defined by a series of split based on the input variables values. You have formal definition of the "shape" of the tree, what is a split, how to define a split criterion.

What you can do is to take your tree model (e.g. that in the link) and give a definition in terms of logical expression:

(Outlook = Sunny AND Humidity = Normal)

OR

(Outlook = Overcast)

OR

(Outlook = Rain AND Wind = Weak)

https://www.projectrhea.org/rhea/images/8/8b/Decision_OldKiwi.jpg