Category » Recursive Partitioning

There were 4 articles found in this category:

1. Interpreting the Correlation/Interaction view
   From the HelixTree Manual: The Correlation Interaction View The figure aboveshows a matrix of the selected variables in the order they were sorted in the Multitree Model window. The numbers appearing in the black diagonal blocks represent the average proportion of observations described by the ...
2. Does the adjusted p-value (Ap) include as possible splits those that do not contain the minimum number of observations in the nodes, as required in the tree options?
   The short answer is no. For tree splits, the factors used in the adjusted p calculations were derived from simulations on random data from the appropriate distribution (binary or normal) where no constraint was placed on the minimum observations in the nodes. Thus, if you have a constraint on th ...
3. How does the random tree creation tool work in HelixTree?
   The random tree creation tool will build random tree models of your data, where the split variables are randomly chosen from a list of significant effects (by default 10 or less). These random trees can then be analyzed in various ways. We can look and see which variables are used across all the ...
4. The meaning of recursive split in HelixTree
   The recursive split will build a dendrogram or tree-like model. The top node is the root node and represents the entire data set. (n=number of observations, u=mean & s=standard deviation). From this root node other branches will grow, forming subgroups, and so on, until a termination criterion i ...