Parameters of a Decision Tree in R

Question

Below is part of the output of a decision tree. Would like to know what does "yval" indicate

node), split, n, deviance, yval
  * denotes terminal node

 1) root 49381 732368600  38.23232  
   2) t1=NEG 7948 126010000 -22.73415  
     4) m1=A,B,C,D,E,F,G,H,I,J,K,L,M,O,P,Q,R,S,T,Z 5845 105874600 -39.38683  
       8) Boo_hospital_is_networked>=0.5 2885  45348390 -57.38094  
        16) m1=B,C,F,G,H,I,J,O,P,Q,Z 1311  19158330 -66.77040 *
        17) m1=A,D,E,K,L,M,R,S,T 1574  25978210 -49.56036 *

Thanks

Answer 1

yval is the predicted response at that node. For example, yval for node 1 (the root) is 38.23, which is the average response value for your training dataset. The values for nodes 16 and 17, the leaves, are -66.77 and -49.56, so these are the predicted values for any observations fallign into these nodes.