Binomial Trees

A binomial tree is a general tree with a very special shape:

Definition (Binomial Tree)  The binomial tree of order with root R is the tree defined as follows

1. If k=0, . I.e., the binomial tree of order zero consists of a single node, R.
2. If k>0, . I.e., the binomial tree of order k>0 comprises the root R, and k binomial subtrees, , , ..., .

Figure  shows the first five binomial trees, - . It follows directly from Definition  that the root of , the binomial tree of order k, has degree k. Since k may arbitrarily large, so too can the degree of the root. Furthermore, the root of a binomial tree has the largest fanout of any of the nodes in that tree.

  
Figure: Binomial Trees , , ...,

The number of nodes in a binomial tree of order k is a function of k:

Theorem  The binomial tree of order k, , contains nodes.

extbfProof (By induction). Let be the number of nodes in , a binomial tree of order k.

Base Case By definition, consists a single node. Therefore .

Inductive Hypothesis Assume that for , for some . Consider the binomial tree of order l+1:

Therefore the number of nodes in is given by

Therefore, by induction on l, for all .

It follows from Theorem  that binomial trees only come in sizes that are a power of two. I.e., . Furthermore, for a given power of two, there is exactly one shape of binomial tree.

Theorem  The height of , the binomial tree of order k, is k.

extbfProof (By induction). Let be the height of , a binomial tree of order k.

Base Case By definition, consists a single node. Therefore .

Inductive Hypothesis Assume that for , for some . Consider the binomial tree of order l+1:

Therefore the height is given by

Therefore, by induction on l, for all .

Theorem  tells us that the height of a binomial tree of order k is k and Theorem  tells us that the number of nodes is . Therefore, the height of is exactly .

Figure  shows that there are two ways to think about the construction of binomial trees. The first way follows directly from the Definition . I.e., binomial consists of a root node to which the k binomial trees , , ..., are attached as shown in Figure  (a).

  
Figure: Two Views of Binomial Tree

Alternatively, we can think of as being comprised of two binomial trees of order k-1. For example, Figure  (b) shows that is made up of two instances of . In general, suppose we have two trees of order k-1, say and , where . Then we can construct a binomial tree of order k by combining the trees to get

Why do we call a binomial tree? It is because the number of nodes at a given depth in the tree is determined by the binomial coefficient. And the binomial coefficient derives its name from the binomial theorem. And the binomial theorem tells us how to compute the power of a binomial . And a binomial is an expression which consists of two terms, such as x+y. That is why it is called a binomial tree!

Theorem (Binomial Theorem)  The power of the binomial x+y for is given by

where is called the binomial coefficient  .

extbfProof The proof of the binomial theorem is left as an exercise for the reader (Exercise ).

The following theorem gives the expression for the number of nodes at a given depth in a binomial tree:

Theorem  The number of nodes at level l in , the binomial tree of order k, where , is given by the binomial coefficient .

extbfProof (By induction). Let be the number of nodes at level l in , a binomial tree of order k.

Base Case Since contains a single node, there is only one level in the tree, l=0, and exactly one node at that level. Therefore, .

Inductive Hypothesis Assume that for , for some . The binomial tree of order h+1 is composed of two binomial trees of height h, one attached under the root of the other. Hence, the number of nodes at level l in is equal to the number of nodes at level l in plus the number of nodes at level l-1 in :

Therefore by induction on h, .

Copyright © 1997 by Bruno R. Preiss, P.Eng. All rights reserved.